1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %
3: % Electronic Transport in DNA --- the Disorder Perspective
4: %
5: % D Klotsa, RA Roemer, MS Turner
6: %
7: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
8: % $Header: /home/phsht/DNA/paper/LongPaper/RCS/paper.tex,v 1.45 2005/04/04 15:17:24 phsht Exp phsht $
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10:
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: % $Log: paper.tex,v $
13: % Revision 1.45 2005/04/04 15:17:24 phsht
14: % some more fidling with figs. 15+16 to make it into the archive large-file-rejection policy
15: %
16: % Revision 1.44 2005/04/04 15:02:35 phsht
17: % q-bio version, with Figs. 15/16 at reduced resolution
18: %
19: % Revision 1.43 2005/04/04 13:57:37 phsht
20: % made into double-spaced single column and also biophysJ style references;
21: % kept the reference numbers, thoughm since no simply style available
22: %
23: % Revision 1.42 2005/04/04 13:31:13 phsht
24: % saline -> ionic
25: %
26: % Revision 1.41 2005/04/04 13:24:51 phsht
27: % I didn't do much in the end. mainly abstract & title and reordered
28: % the introduction a little. If you are happy we can send it.
29: %
30: % Revision 1.40 2005/03/21 14:40:34 phsht
31: % long list of KDK corrections as in sent .PDF
32: %
33: % Revision 1.39 2005/03/17 16:35:17 phsht
34: % added some more on the 1D TB models
35: %
36: % Revision 1.38 2005/03/17 16:25:05 phsht
37: % worked on the aX references and completed them
38: %
39: % Revision 1.37 2005/03/17 15:54:22 phsht
40: % replaced XXX by MutY and included the relevant reference
41: %
42: % Revision 1.36 2005/03/17 12:22:08 phsht
43: % changed the discussion to take into account the new binary figures
44: %
45: % Revision 1.35 2005/03/14 23:16:03 phsht
46: % changed all binary figures to correctly reflect the disorder values \pm W/2;
47: % addeda horizontal line at K=100 for the VII plots (Fig. 15)
48: %
49: % Revision 1.34 2005/03/14 17:03:39 phsht
50: % small changes
51: %
52: % Revision 1.33 2005/03/14 09:27:40 phsht
53: % small but many typo corrections and consistencies
54: %
55: % Revision 1.32 2005/03/10 17:13:24 phsht
56: % included the new DNA figure
57: %
58: % Revision 1.31 2005/03/09 16:01:00 phsht
59: % minor changes to MST version to include the bib and new citations
60: %
61: % Revision 1.30 2005/03/09 15:51:48 phsht
62: % edited by MST
63: %
64: % Revision 1.30 2005/03/09 06:20:00 phscz
65: % edited by MT
66: %
67: % Revision 1.29 2005/03/06 18:13:28 phsht
68: % updated a1, a8, a9 according to DKs suggestions
69: %
70: % Revision 1.28 2005/03/05 19:08:18 phsht
71: % e.g.i.e. replaced
72: %
73: % Revision 1.27 2005/03/03 19:36:07 phsht
74: % DK grammar corrections
75: %
76: % Revision 1.26 2005/02/23 18:24:08 phsht
77: % added remarks on the DOS and finished the conclusions (draft)
78: %
79: % Revision 1.25 2005/02/23 16:22:49 phsht
80: % colloquium
81: %
82: % Revision 1.24 2005/02/23 12:10:05 phsht
83: % this version sent to DK/MT
84: %
85: % Revision 1.23 2005/02/22 16:55:44 phsht
86: % spell checked
87: %
88: % Revision 1.22 2005/02/22 16:41:36 phsht
89: % went thru the reference and updated the bib file, some {a}'s are still to be added
90: %
91: % Revision 1.21 2005/02/22 10:56:31 phsht
92: % all labels corrected
93: %
94: % Revision 1.20 2005/02/21 18:12:00 phsht
95: % added the 1st bits on the window technique
96: %
97: % Revision 1.19 2005/02/21 10:43:41 phsht
98: % remade the figures to include the DOS information
99: %
100: % Revision 1.18 2005/02/11 10:43:57 phsht
101: % added DKs correlation definition
102: %
103: % Revision 1.17 2005/02/10 16:10:02 phsht
104: % up to WINDOW
105: %
106: % Revision 1.16 2005/02/10 10:05:39 phsht
107: % added discussion as to why delocalization happens
108: %
109: % Revision 1.15 2005/02/09 22:13:01 phsht
110: % up to explaining the reason for the delocalization
111: %
112: % Revision 1.14 2005/02/09 18:21:37 phsht
113: % going home
114: %
115: % Revision 1.13 2005/02/09 17:17:56 phsht
116: % included new BW figures for section on disordered DNA
117: %
118: % Revision 1.12 2005/02/08 21:36:17 phsht
119: % from home
120: %
121: % Revision 1.11 2005/02/08 17:53:02 phsht
122: % converted the first two figures into BW
123: %
124: % Revision 1.10 2005/02/08 15:50:05 phsht
125: % Sokal talk
126: %
127: % Revision 1.9 2005/02/07 17:09:52 phsht
128: % going home
129: %
130: % Revision 1.8 2005/02/07 14:04:17 phsht
131: % project students
132: %
133: % Revision 1.7 2005/02/07 12:30:00 phsht
134: % lunch
135: %
136: % Revision 1.6 2005/02/04 17:56:20 phsht
137: % until the beginning of "Results for ordered"
138: %
139: % Revision 1.5 2005/02/02 16:28:19 phsht
140: % physics colloquium
141: %
142: % Revision 1.4 2005/02/01 18:08:55 phsht
143: % squash with matt
144: %
145: % Revision 1.3 2005/01/31 15:52:46 phsht
146: % reworking the introduction
147: %
148: % Revision 1.2 2005/01/28 18:10:59 phsht
149: % going home
150: %
151: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
152:
153: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
154: % FRONTMATTER
155: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
156:
157: \documentclass[pra,twocolumn,a4paper,showpacs,floatfix]{revtex4} % double column
158: %\documentclass[pra,preprint,a4paper,showpacs]{revtex4} % single column
159: %\documentclass[12pt]{article} % single column
160:
161: \usepackage{graphicx}
162: \usepackage{amsmath,amssymb}
163: %\usepackage{showlabels}
164: %\usepackage{datetime}
165: %\usepackage{times}
166: \newcommand{\figwidth}{0.95\columnwidth}
167:
168: \begin{document}
169:
170: \title{Electronic Transport in DNA}
171:
172: \author{Daphne Klotsa, Rudolf A.\ R\"{o}mer, Matthew S.\ Turner}
173: %
174: \affiliation{Physics Department and Centre for Scientific
175: Computing, University of Warwick, Coventry CV4 7AL, U.K.}
176:
177: %\date{$Revision: 1.45 $, compiled \today, \currenttime}
178: \date{$Revision: 1.45 $, compiled \today}
179:
180: \begin{abstract}
181: We study the electronic properties of DNA by way of a tight-binding
182: model applied to four particular DNA sequences. The charge transfer
183: properties are presented in terms of localisation lengths, crudely speaking the length over which electrons travel. Various types of disorder, including random potentials, are employed to account for different real
184: environments. We have performed calculations on poly(dG)-poly(dC),
185: telomeric-DNA, random-ATGC DNA and $\lambda$-DNA. We find that random
186: and $\lambda$-DNA have localisation lengths allowing for electron
187: motion among a few dozen base pairs only. A novel enhancement of
188: localisation lengths is observed at particular energies for an
189: increasing binary backbone disorder. We comment on the possible
190: biological relevance of sequence dependent charge transfer in DNA.
191: \end{abstract}
192:
193: \pacs{72.15.Rn, 87.15.Cc, 73.63.-b}
194:
195: \maketitle
196:
197: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
198: \section{Introduction}
199: \label{sec-introduction}
200: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
201:
202: The question of whether DNA conducts electric charges is intriguing to
203: physicists and biologists alike. The suggestion that electron transfer/transport in DNA might be
204: biologically important has triggered a series of experimental and
205: theoretical investigations
206: \cite{MurAJG93,BerBR00,WanFSB00,DelB03,OneDB04,EndCS04}. Processes that
207: possibly use electron transfer include the function of DNA damage
208: response enzymes, transcription factors or polymerase co-factors all of
209: which play important roles in the cell \cite{AlbBLR94}. Indeed there is
210: direct evidence \cite{BooLCD03} that MutY --- a DNA base excision repair
211: enzyme with an [4Fe4S]$^+$ cluster of undetermined function ---
212: %name this molecule
213: takes part in some kind of electron transfer as part of the DNA repair
214: process \cite{OneF93,RetHBL93}. This seems consistent with studies in
215: which an electric current is passed through DNA revealing that damaged
216: regions have significantly different electronic behaviour than healthy
217: ones \cite{BooLCD03}.
218:
219: For physicists, the continuing progress
220: of nanotechnologies and the consequent need for further size
221: miniaturisation makes the DNA molecule an excellent candidate for
222: molecular electronics \cite{CunCPD02,GarABG01,RakAPK01,BhaBB03}. DNA
223: might serve as a wire, transistor, switch or rectifier depending on its
224: electronic properties \cite{DekR01,PorCD04,EndCS04}.
225:
226: In its natural environment, DNA is always in liquid solution and
227: therefore experimentally one can study the molecule either in solution
228: or in artificially imposed dry environments. In solution experiments DNA
229: can be chemically processed to host a donor and an acceptor molecule at
230: different sites along its long axis. Photo-induced charge transfer rates
231: can then be measured whilst the donor/acceptor molecules, the distance
232: and the sequence of DNA that lies between them are varied. The
233: reactions are observed to depend on the type of DNA used, the
234: intercalation, the integrity of the intervening base pair stack and,
235: albeit weakly, on the molecular distance
236: \cite{BerBR00,OneDB04,BooLCD03,DelB03,TreHB01}.
237:
238: Direct conductivity measurements on dry DNA have also been preformed in
239: the past few years. The remarkable diversity that characterises the
240: results seems to arise from the fact that many factors need to be
241: experimentally controlled. These include methods for DNA alignment and
242: drying, the nature of the devices used to measure the conductivity, the
243: type of metallic contacts and the sequence and length of the DNA. DNA
244: has been reported to be an insulator \cite{BraESB98}, an ohmic conductor
245: \cite{RakAPK01,FinS99,NakGSO03,AsaT00,OkaKTS98} and a semiconductor
246: \cite{PorBVD00}. Theoretically, single-step super exchange
247: \cite{MurAJG93} and multi-step hopping \cite{BixGWL99} models have
248: provided interpretations of solution experiments. For experiments in dry
249: DNA, several additional approaches such as variable range hopping
250: \cite{YuS01}, one-dimensional quantum mechanical tight-binding models
251: \cite{CunCPD02,Roc03,ZhaU04a,ZhaU04b,RocBMK03,WanLS04} and non-linear
252: methods \cite{CueS04,Pey04} have also been proposed.
253: %need to say something like non linear phonon approaches..
254:
255: Despite the lack of a consistent picture for the electronic properties
256: of DNA, one conclusion has been established: the environment of the DNA
257: impacts upon its structural, chemical and thus probably also electronic
258: properties. Both theoretical and experimental studies show that the
259: temperature and the type of solution surrounding DNA have a significant
260: effect on its structure and shape \cite{YuS01,BarCJL01,BruGOR00}. The
261: effect of the environment is a key one to this report, where the
262: environmental fluctuations are explicitly modelled as providing
263: different types of disorder.
264:
265: In this work, we focus on whether DNA, when treated as a quantum wire in
266: the fully coherent low-temperature regime, is conducting or not. To this
267: end, we study and generalise a tight-binding model of DNA which has been
268: shown to reproduce experimental \cite{CunCPD02} as well as {\em
269: ab-initio} results \cite{DavI04}. A main feature of the model is the
270: presence of sites which represent the sugar-phosphate backbone of DNA
271: but along which no electron transport is permissible. We measure the
272: ``strength'' of the electronic transport by the {\em localisation
273: length} $\xi$, which roughly speaking parametrises whether an electron
274: is confined to a certain region $\xi$ of the DNA (insulating behaviour)
275: or can proceed across the full length $L$ ($\leq \xi$) of the DNA
276: molecule (metallic behaviour).
277:
278: Sections \ref{sec-models}--\ref{sec-localization} introduce our models
279: and the numerical approach. In section \ref{sec-results-clean}, we
280: show that DNA sequences with different arrangement of nucleotide bases
281: Adenine (A), Cytosine (C), Guanine (G) and Thymine (T) exhibit different
282: $\xi$'s when measured, e.g.\ as function of the Fermi energy $E$. The
283: influence of external disorder, modelling variants in the solution,
284: bending of the DNA molecule, finite-temperature effects, etc., is
285: studied in section \ref{sec-results-disordered} where we show that,
286: surprisingly, the models support an increase of $\xi$ when disorder is
287: increased. We explain that this effect is linked to the existence of the
288: backbone sites.
289:
290: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
291: \section{Tight-binding models for DNA with a gap in the spectrum}
292: \label{sec-models}
293: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
294:
295: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
296: \subsection{The Fishbone model}
297: \label{sec-fishbone}
298:
299: DNA is a macro-molecule consisting of repeated stacks of bases formed by
300: either AT (TA) or GC (CG) pairs coupled via hydrogen bonds and held in
301: the double-helix structure by a sugar-phosphate backbone. In Fig.\
302: \ref{fig-DNA}, we show a schematic drawing.
303: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
304: \begin{figure}
305: \centering
306: % \includegraphics[width=\figwidth]{DNA-mine-bw.eps}
307: \includegraphics[width=\figwidth]{fig/fig-DNA-BW.eps}
308: \caption{\label{fig-DNA}
309: The chemical composition of DNA with the four bases Adenine,
310: Thymine, Cytosine, Guanine and the backbone. The backbone is made of
311: phosphorylated sugars shown in yellow and brown.}
312: % check colours here
313: \end{figure}
314: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
315: %
316: In most models of electronic transport \cite{CunCPD02,Zho03} it has been
317: assumed that the transmission channels are along the long axis of the
318: DNA molecule \cite{perpendiculartolongaxis} and that the conduction path
319: is due to $\pi$-orbital overlap between consecutive bases
320: \cite{TreHB01}; density-functional calculations \cite{PabMCH00} have
321: shown that the bases, especially Guanine, are rich in $\pi$-orbitals.
322: Quantum mechanical approaches to the problem mostly use strictly
323: one-dimensional (1D) tight-binding models
324: \cite{Roc03,ZhaU04a,ZhaU04b,RocBMK03,WanLS04}.
325:
326: Of particular interest to us is a quasi-1D model \cite{CunCPD02} which
327: includes the backbone structure of DNA explicitly and exhibits a
328: semiconducting gap.
329: %
330: This {\em fishbone model}, shown in Fig.\ \ref{fig-fishbone}, has one
331: central conduction channel in which individual sites represent a
332: base-pair; these are interconnected and further linked to upper and
333: lower sites, representing the backbone, but are \emph{not}
334: interconnected along the backbone. Every link between sites implies the
335: presence of a hopping amplitude. The Hamiltonian for the fishbone model
336: $(H_F)$ is given by:
337: \begin{eqnarray}
338: H_F &=& \sum_{i=1}^{L}
339: \sum_{q=\uparrow,\downarrow} \left(
340: -t_{i} |i \rangle \langle i+1|-t_i^q |i,q \rangle \langle i|
341: \right.
342: \nonumber \\
343: & &
344: \left.+ \varepsilon_i |i \rangle \langle i| + \varepsilon_i^q |i,q
345: \rangle \langle i,q| \right) + h.c. \label{eq-ham1D}\label{eq-fishbone}
346: \end{eqnarray}
347: where $t_{i}$ is the hopping between nearest-neighbour sites $i,i+1$
348: along the central branch, $t_i^q$ with $q=\uparrow, \downarrow$ gives
349: the hopping from each site on the central branch to the upper and lower
350: backbone respectively. Additionally, we denote the onsite energy at each
351: site along the central branch by $\varepsilon_i$ and the onsite energy
352: at the sites of the upper and lower backbone is given by
353: $\varepsilon_i^q$, with $q=\uparrow\downarrow$. $L$ is the number of
354: sites/bases in the sequence.
355: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
356: \begin{figure}
357: \centering
358: \includegraphics[width=\figwidth]{fishbone-model-grey.eps}
359: \caption{\label{fig-fishbone}
360: The fishbone model for electronic transport along DNA corresponding
361: to the Hamiltonian given in Eq.\ (\ref{eq-ham1D}). Lines denote
362: hopping amplitudes and circles give the central (grey) and backbone
363: (open) sites.}
364: \end{figure}
365: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
366: %
367: The model (\ref{eq-fishbone}) clearly represents a dramatic
368: simplification of DNA. Nevertheless, in Ref.\ \cite{CunCPD02} it had
369: been shown that this model when applied to an artificial sequence of
370: repeated GC base pairs, poly(dG)-poly(dC) DNA, reproduces experimental
371: data current-voltage measurements when $t_{i}=0.37 e$V and
372: $t_i^q=0.74 e$V are being used. Therefore, we will assume $t_i^q = 2
373: t_{i}$ and set the energy scale by $t_{i}\equiv 1$ for hopping
374: between GC pairs. In what follows we will adopt energy
375: units in which $eV=1$ throughout.
376:
377: For natural DNA sequences, we need to know how the hopping amplitudes
378: vary as the electron moves between like pairs, i.e.\ from GC to GC or
379: from AT to AT, and unlike pairs, i.e., from GC to AT and vice versa. We
380: choose $t_{i}=1$ between identical and matching bases (e.g.\ AT/TA,
381: GC/CG). Assuming that the wavefunction overlap between consecutive bases
382: along the DNA strand is weaker between unlike and non-matching bases
383: (AT/GC, TA/GC, etc.) we thus choose $1/2$.
384:
385:
386: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
387: \subsection{The Ladder model}
388: \label{sec-ladder}
389:
390: We performed semi-empirical calculations on DNA base pairs and stacks
391: using the SPARTAN quantum chemistry software package \cite{Spartan}. The
392: results have shown that the relevant electronic states of DNA
393: (highest-occupied and lowest-unoccupied molecular orbitals with and
394: without an additional electron) are localised on one of the bases of a
395: pair only. The reduction of the DNA base-pair architecture into a single
396: site per pair, as in the fishbone model (\ref{eq-fishbone}), is
397: obviously a highly simplified approach. As an improvement on this we
398: model each base as a distinct site where the base pair is then weakly
399: coupled by the hydrogen bonds. The resulting 2-channel model is shown in
400: Fig.\ \ref{fig-ladder}. This {\em ladder} model is a planar projection
401: of the structure of the DNA with its double-helix unwound. We note that
402: results for electron transfer also suggest that the transfer proceeds
403: preferentially down one strand \cite{KelB99}.
404: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
405: \begin{figure}
406: \centering
407: % \includegraphics[width=\figwidth]{LADDER_COLOUR.eps}
408: \includegraphics[width=\figwidth]{ladder-model-grey.eps}
409: \caption{\label{fig-ladder}
410: The ladder model for electronic transport along DNA. The model
411: corresponds to the Hamiltonian (\ref{eq-ladder}).}
412: \end{figure}
413: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
414: There are two central branches, linked with one another, with
415: interconnected sites where each represents a complete base and which are
416: additionally linked to the upper and lower backbone sites. The backbone
417: sites as in the fishbone model are not interconnected. The Hamiltonian
418: for the ladder model is given by
419: \begin{eqnarray}
420: H_{L} &=& \sum_{i=1}^{L} \left[
421: \sum_{\tau=1,2}
422: \left( t_{i,\tau}|i,\tau\rangle \langle i+1,\tau| +
423: \varepsilon_{i,\tau} |i,\tau\rangle \langle i,\tau| \right) \right.
424: \nonumber \\
425: & & \mbox{} + \sum_{q=\uparrow,\downarrow}
426: \left( t_i^q |i,\tau\rangle \langle i,q(\tau)|+
427: \varepsilon_i^q|i,q\rangle \langle i,q| \right)
428: \nonumber \\
429: & & \mbox{ }
430: + t_{1,2}|i,1\rangle \langle i,2|
431: \Big]
432: + h.c. \label{eq-ham2D}\label{eq-ladder}
433: \end{eqnarray}
434: where $t_{i,\tau}$ is the hopping amplitude between sites along each
435: branch $\tau=1$, $2$ and $\varepsilon_{i,\tau}$ is the corresponding
436: onsite potential energy. $t_i^q$ and and $\varepsilon_i^q$ as before
437: give hopping amplitudes and onsite energies at the backbone sites. Also,
438: $q(\tau)=\uparrow, \downarrow$ for $\tau=1, 2$, respectively. The new
439: parameter $t_{12}$ represents the hopping between the two central
440: branches, i.e., perpendicular to the direction of conduction. SPARTAN
441: results suggest that this value, dominated by the wave function overlap
442: across the hydrogen bonds, is weak and so we choose $t_{12}= 1/10$.
443:
444: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
445: \subsection{Including disorder}
446: \label{sec-disorder}
447:
448: In order to study the transport properties of DNA, we could now either
449: use artificial DNA (poly(dG)-poly(dC) \cite{PorBVD00}, random sequences
450: of A,T,G,C \cite{PenBGH92,YamSHA04}, etc.) or natural DNA (bacteriophage
451: $\lambda$-DNA \cite{PabMCH00}, etc.). The biological content of the
452: sequence would then simply be encoded in a specific sequence of hopping
453: amplitudes $1$ and $1/2$ between like and unlike base-pair sequences.
454: However, in vivo and most experimental situations, DNA is exposed to
455: diverse environments and its properties, particularly those related to
456: its conformation, can change drastically depending on the specific
457: choice. The solution, thermal effects, presence of binding and
458: packaging proteins and the available space are factors that alter the
459: structure and therefore the properties that one is measuring
460: \cite{YuS01,BarCJL01}. Clearly, such dramatic changes should also be
461: reflected in the electronic transport characteristics. Since it is
462: precisely the backbone that will be most susceptible to such influences,
463: we model such environmental fluctuations by including variations in the
464: onsite potentials $\varepsilon_{i,q}$.
465:
466: Different experimental situations will result in a different
467: modification of the backbone electronic structure, and we model this by
468: choosing different distribution functions for the onsite potentials,
469: ranging from uniform disorder $\varepsilon_{i,q}\in [-W/2,W/2]$, to
470: Gaussian disorder and on to binary disorder $\varepsilon_{i,q}=
471: \pm W/2$. $W$ is a measure for the strength of the disorder in all
472: cases. Particularly the binary disorder model can be justified by the
473: localisation of ions or other solutes at random positions
474: along the DNA strand \cite{BarCJL01}.
475:
476: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
477: \subsection{Effective models and the energy gap}
478: \label{sec-effective}
479:
480: Due to the non-connectedness of the backbone sites along the DNA
481: strands, the models (\ref{eq-fishbone}) and (\ref{eq-ladder}) can be
482: further simplified to yield models in which the backbone sites are
483: incorporated into the electronic structure of the DNA. The effective
484: fishbone model is then given by
485: \begin{eqnarray}
486: \tilde{H}_F &=& \sum_{i=1}^{L}
487: -t_{i} |i \rangle \langle i+1| + h.c.
488: \nonumber \\
489: & & \mbox{ } + \left[\varepsilon_i -
490: \sum_{q=\uparrow,\downarrow}
491: \frac{\left(t_{i}^{q}\right)^{2}}{\varepsilon_{i}^{q} - E}\right]
492: |i \rangle \langle i| \quad .
493: \label{eq-fishbone-effective}
494: \end{eqnarray}
495: Similarly, the effective ladder model reads as
496: \begin{eqnarray}
497: \tilde{H}_{L} &=& \sum_{i=1}^{L}
498: t_{1,2}|i,1\rangle \langle i,2| +
499: \sum_{\tau=1,2}
500: t_{i,\tau}|i,\tau\rangle \langle i+1,\tau|
501: \nonumber \\
502: & &
503: +
504: \left[ \varepsilon_{i,\tau} -
505: \frac{\left(t_{i}^{q(\tau)}\right)^{2}}{\varepsilon_{i}^{q(\tau)} - E}
506: \right]
507: |i,\tau\rangle \langle i,\tau|
508: \nonumber \\
509: & & \mbox{ }+ h.c. \quad . \label{eq-ladder-effective}
510: \end{eqnarray}
511: In these two models, the backbone has been incorporated into an
512: energy-dependent onsite potential on the main DNA sites. This
513: re-emphasises that the presence of the backbone influences the local
514: electronic structure on the DNA bases and similarly, any variation in
515: the backbone disorder potentials $\varepsilon_{i}^{\uparrow,\downarrow}$
516: will results in a variation of {\em effective} onsite potentials as
517: given in the brackets of Eqs.\ (\ref{eq-fishbone-effective}) and
518: (\ref{eq-ladder-effective}).
519:
520: Both models allow to quickly calculate the gap of the completely ordered
521: system (all onsite potentials zero) by assuming that the lowest-energy
522: state $\psi=\sum_{i} \psi_{i(,\tau)} |i(,\tau)\rangle$ in each band
523: corresponds to constant $\psi_i$ ($\psi_{i,\tau}$) whereas for the
524: highest-energy states, a checker-board pattern is obtained with
525: $\psi_{i}=\psi_{i+1}$ ($\psi_{i,\tau}=-\psi_{i+1,\tau}$,
526: $\psi_{i,1}=-\psi_{i,2}$). For the fishbone model, this shows that,
527: e.g.\ $E_{{\rm min},\mp}=-
528: t_{i}\mp\sqrt{t_{i}^2+t_{i,\uparrow}^2+t_{i,\downarrow}^{2}}$ and
529: $E_{{\rm max},\mp}=
530: t_{i}\mp\sqrt{t_{i}^2+t_{i,\uparrow}^2+t_{i,\downarrow}^{2}}$.
531: %
532: For the chosen set of hopping parameters for
533: (\ref{eq-fishbone-effective}) and (\ref{eq-ladder-effective}), this
534: gives $E_{{\rm min},\mp}= -4, 2$ and $E_{{\rm max},\mp}= -2, 4$ for the
535: fishbone model and $E_{{\rm min},\mp}\approx -3.31, 1.21$ and $E_{{\rm
536: max},\mp}= -1.21, 3.31$ for the ladder model.
537:
538: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
539: \section{The numerical approach and localisation}
540: \label{sec-localization}
541: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
542:
543: There are several approaches suitable for studying the transport
544: properties of the models (\ref{eq-fishbone}) and (\ref{eq-ladder}) and
545: these can be found in the literature on transport in solid state
546: devices, or, perhaps more appropriately, quantum wires. Since the
547: variation in the sequence of base pairs precludes a general solution, we
548: will use two methods well-known from the theory of disordered systems
549: \cite{RomS03}.
550:
551: The first method is the iterative transfer-matrix method (TMM)
552: \cite{PicS81a,PicS81b,MacK83,KraM93,Mac94} which allows us in principle
553: to determine the localisation length $\xi$ of electronic states in
554: systems with cross sections $M=1$ (fishbone) and $2$ (ladder) and length
555: $L \gg M$, where typically a few million sites are needed for $L$ to
556: achieve reasonable accuracy for $\xi$. However, in the present
557: situation we are interested in finding $\xi$ also for viral DNA strands
558: of typically only a few ten thousand base-pair long sequences. Thus in
559: order to restore the required precision, we have modified the conventional
560: TMM and now perform the TMM on a system of fixed length $L_0$. This
561: modification has been previously used \cite{FraMPW95,RomS97b,NdaRS04}
562: and may be summarised as follows:
563: %
564: After the usual forward calculation with a global transfer matrix ${\cal
565: T}_{L_0}$, we add a backward calculation with transfer matrix ${\cal
566: T}^{\rm b}_{L_0}$. This forward-backward-multiplication procedure is
567: repeated $K$ times. The effective total number of TMM multiplications is
568: $L_{\rm }=2KL_0$ and the global transfer-matrix is ${\tau}_{L_{\rm }} =
569: \left( {\cal T}^{\rm b}_{L_0} {\cal T}_{L_0}\right)^K$. It can be
570: diagonalised as for the standard TMM with $K\rightarrow \infty$ to give
571: ${\tau}^{\dagger}_{L_{\rm }} {{\tau}_{L_{\rm }}} \rightarrow \exp[ {\rm
572: diag}(4KL_0/\xi_{\tau})]$ with $\tau=1$ or $\tau= 1, 2$ for fishbone
573: and ladder model, respectively. The largest $\xi_{\tau} \forall \tau$
574: then corresponds to the localisation lengths of the electron on the DNA
575: strand and will be measured in units of the DNA base-pair spacing
576: ($0.34$ nm).
577:
578: The second method that we will use is the recursive Green function
579: approach pioneered by MacKinnon \cite{Mac80,Mac85}. It can be used to
580: calculate the dc and ac conductivity tensors and the density of states
581: (DOS) of a $d$-dimensional disordered system and has been adopted to
582: calculate all kinetic linear-transport coefficients such as
583: thermoelectric power, thermal conductivity, Peltier coefficient and
584: Lorentz number \cite{RomMV03}.
585:
586: The main advantage of both methods is that they work reliably (i) for
587: short DNA strands ranging from 13 (DFT studies \cite{PabMCH00}) base
588: pairs up to 30 base pairs length which are being used in the nanoscopic
589: transport measurements \cite{DavI04} as well as (ii) for somewhat longer
590: DNA sequences as modelled in the electron transfer results and (iii)
591: even for complete DNA sequences which contain, e.g.\ for human
592: chromosomes up to 245 million base pairs \cite{AlbBLR94}.
593:
594: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
595: \section{DNA sequences}
596: \label{sec-dna}
597: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
598:
599: The exact arrangement of the four bases A, T, G, C determines the nature
600: and function of its associated DNA strand such as the chemical
601: composition of the proteins which are encoded. While previous studies
602: have aimed to elucidate whether DNA conducts at all, we shall also focus
603: our attention to investigate how different DNA sequences, be they
604: artificial or naturally occurring, ``conduct'' charge differently. Thus
605: we study a set of different DNA.
606:
607: A convenient starting point for most electronic transport studies
608: \cite{PorCD04} is the aforementioned poly(dG)-poly(dC) sequence, which
609: corresponds to a simple repetition of a GC (or CG) pair. Note that
610: within our models, there is no difference between GC and CG pairs.
611: Although not occurring naturally, such sequences can be synthesised
612: easily. Another convenient choice of artificial DNA strand is a simple
613: {\em random} sequence of the four bases, which we construct with equal
614: probability for all 4 bases. However, they are not normally used in
615: experiments.
616:
617: As DNA samples existing in living organisms, we shall use $\lambda$-DNA
618: of the bacteriophage virus \cite{lambda} which has a sequence of 48502
619: base pairs. It corresponds to a bacterial virus and is biologically very
620: well characterised. We also investigate the $29728$ bases of the SARS
621: virus \cite{sars}.
622: %
623: Telomeric DNA is a particular buffer part at the beginning and ends of
624: of DNA strands for eukaryote cells \cite{AlbBLR94}. In mammals it is a
625: Guanine rich sequence in which the pattern TTAGGG is repeated over
626: thousands of bases. Its length is known to vary widely between species
627: and individuals but we assume a length of 6000 base-pairs.
628: %
629: Last, we show some studies of centromeric DNA for chromosome 2 of yeast
630: with 813138 base pairs \cite{cen2}. This DNA is also reportedly rich in G
631: bases and has a high rate of repetitions which should be favourable for
632: electronic transport.
633:
634: Initially, we will compute transport properties for complete DNA
635: sequences, i.e.\ including and not differentiating between coding and
636: non-coding sequences (this distinction applies to the naturally
637: occurring DNA strands only). However, we will later also study the
638: difference between those two different parts of a given DNA. We
639: emphasise that while non-coding DNA suffers from the label of ``junk'',
640: it is now known to play several important roles in the functioning of
641: DNA \cite{AlbBLR94}.
642:
643: Before leaving the description of our DNA sequences, we note that
644: occasionally, we show results for ``scrambled'' DNA. This is DNA with
645: the same number of A, T, C, G bases, but with their order randomised.
646: Clearly, such sequences contain the same set of electronic potentials
647: and hopping variations, but would perform quite differently if released
648: into the wild. A comparison of their transport properties with those
649: from the original sequence thus allows to measure how important the
650: exact fidelity of a sequence is.
651:
652: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
653: \section{Results for clean DNA}
654: \label{sec-results-clean}
655: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
656:
657: Let us start by studying the localisation properties of DNA without any
658: onsite disorder either at $\varepsilon_{i,\tau}$ or at
659: $\varepsilon_{i,q}$. For a poly(dG)-poly(dC) sequence, both fishbone
660: and ladder model produce two separate energy bands between the extremal
661: values computed at the end of section \ref{sec-effective}. Within these
662: energy bands, the electronic states are extended with infinite
663: localisation length $\xi$ as expected. Outside the bands, transport is
664: exponentially damped due to an absence of states and the $\xi$ values
665: are very close the zero. In Fig.\ \ref{fig-inverse2d} the
666: resulting {\em inverse} localisation lengths are shown.
667: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
668: \begin{figure}
669: \centering
670: % \includegraphics[width=\figwidth]{inverse_energy_2d_new.eps}
671: \includegraphics[width=\figwidth]{fig/fig-LM-invloc_energy.eps}
672: \caption{\label{fig-inverse2d}
673: Plot of the inverse localisation lengths $\xi$ as a function of
674: Fermi energy for the ladder model (\ref{eq-ladder-effective} and
675: four DNA sequences as well as for the fishbone model with a
676: poly(dG)-poly(dC) sequence. The data for telomeric DNA has been
677: shaded for clarity. Lines are guides to the eye only.}
678: \end{figure}
679: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
680: These are zero for the extended states in the two bands, but finite
681: outside, showing the quick decrease of the localisation lengths outside
682: the bands.
683: %
684: In Fig.\ \ref{fig-energy2d}, we show the same data but now plot the
685: localisation length itself.
686: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
687: \begin{figure}
688: \centering
689: \includegraphics[width=\figwidth]{fig/fig-LM-loc_energy.eps}
690: \caption{\label{fig-energy2d}
691: Localisation lengths as a function of energy for poly(dG)-poly(dC),
692: telomeric, random-ATGC, and $\lambda$-DNA as described in the text.
693: The spectrum is symmetric in energy. The data for telomeric DNA has
694: been shaded for clarity. Lines are guides to the eye only.}
695: \end{figure}
696: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
697: We see that the energy gap observed previously \cite{CunCPD02} for the
698: poly(dG)-poly(dC) sequence in the fishbone model remains. The difference
699: with respect to the ladder model is a slight renormalisation of the gap
700: width.
701: %
702: The localisation lengths of poly(dG)-poly(dC) DNA tend to infinity,
703: meaning that the sequence is perfectly conducting. This is expected due
704: to its periodic electronic structure.
705:
706: Turning our attention to the other three DNA sequences, we find that
707: telomeric DNA also gives rise to perfect conductivity like
708: poly(dG)-poly(dC) DNA. But due to its structure of just 6 repeating base
709: pairs, there is a further split of each band into 3 separate sub-bands.
710: They may be calculated like in section \ref{sec-effective}.
711: %
712: We would like to point out that it may therefore be advantageous to use
713: the naturally occurring telomeric parts of DNA sequences as prime,
714: in-vivo candidates when looking for good conductivity in a DNA strand.
715: %
716:
717: The structure of the energy dependence for the random-ATGC and the
718: $\lambda$-DNA is very different from the preceding two sequences, but it
719: is quite similar between just these two. The biological content of the
720: DNA sequences is --- within the description by our quantum models ---
721: just a sequence of binary hopping elements between like and unlike base
722: pairs. Thus the models are related to the physics of random hopping
723: models \cite{EilRS98a,BisCRS00} and in agreement with these, we see a
724: Dyson peak \cite{Dys53} in the centre of each sub-band. Furthermore, we
725: see that the range of energies for which we observe non-zero
726: localisation lengths is increased into the gap and for large absolute
727: values of the energy. This is similar to the broadening of the single
728: energy band for the Anderson model of localisation \cite{RomS03}.
729: %
730: The localisation lengths, which roughly equal the average distance an
731: electron would be able to travel (conduct), are close to the distance of
732: $20$ bases within the band, with a maximum of $\sim 30$ bases at the
733: centre of each band. Note that this result is surprisingly good ---
734: given the level of abstraction used in the present models --- when
735: compared to the typical distances over which electron transfer processes
736: have been shown to be relevant
737: \cite{WanFSB00,BooLCD03,KelB99,MurAJG93,OneDB04,DelB03,TreHB01}.
738:
739: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
740: \section{Results for disordered DNA}
741: \label{sec-results-disordered}
742: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
743:
744: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
745: \subsection{DNA randomly bent or at finite temperatures}
746: \label{sec-uniform_energy}
747:
748: As argued before, environmental influences on the transport properties
749: of DNA are likely to influence predominantly the electronic structure of
750: the backbone. Within our models, this can be captured by adding a
751: suitable randomness onto the backbone onsite potentials
752: $\varepsilon_{i}^{q}$. In this fashion, we can model for example the
753: influence of a finite-temperature \cite{BruGOR00} and thus a coupling to
754: phonons \cite{GutMC04}. We emphasise however, that in order for our
755: localisation results --- which rely on quantum mechanical interference
756: effects --- to remain valid, the phase breaking lengths should stay much
757: larger than the sequence lengths. Thus the permissible temperature range
758: is a few K only.
759: %
760: The bending of DNA is another possibility which can be modelled by a
761: local, perhaps regular, change in $\varepsilon_{i}^{q}$ along the strand.
762: %
763: Another important aspect is the change in $\varepsilon_{i}^{q}$ due to
764: the presence of a solution in which DNA is normally immersed.
765:
766: All these effects can be modelled in a first attempt by choosing an
767: appropriate distribution function $P(\varepsilon_{i}^{q}$). Let us first
768: choose uniform disorder with $\varepsilon_{i}^{q} \in [-W/2,W/2]$.
769: %
770: In Fig.\ \ref{fig-LM-uniW1-loc_energy} we show the results for all 4 DNA
771: sequences as a function of energy for $W=1$. Comparing this to Fig.\
772: \ref{fig-energy2d}, we see that now all localisation lengths are finite;
773: poly(dG)-poly(dC) and telomeric DNA having localisation lengths of a few
774: hundreds and a few tens of bases, respectively. The localisation lengths
775: for random-ATGC and $\lambda$-DNA are only slightly reduced. In all
776: cases, the structure of 2 energy bands remains. Furthermore, $W=1$
777: already represents a sizable broadening of about $1/2$ the width of each
778: band. Thus although the localisation lengths are finite compared to the
779: results of section \ref{sec-results-clean}, they are still larger than
780: the lengths of the DNA strands used in the nano-electric experiments,
781: implying finite conductances. We remark that the Dyson peaks have
782: vanished as expected \cite{EilRS98a}. We plot the DOS for $\lambda$-DNA
783: in Fig.\ \ref{fig-LM-uniW1-loc_energy} which clearly indicates the $2$
784: bands.
785: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
786: \begin{figure}
787: \centering
788: % \includegraphics[width=\figwidth]{all_seq_BB1_uniform.eps}
789: \includegraphics[width=\figwidth]{fig/fig-LM-uniW1-loc-DOS_energy.eps}
790: \caption{\label{fig-LM-uniW1-loc_energy}
791: Top: Energy dependence of the localisation lengths, $\xi(E)$, for
792: poly(dG)-poly(dC), telomeric, random-ATGC and $\lambda$-DNA in the
793: presence of {\em uniform} backbone disorder with $W=1$. Only every
794: 2nd and 5th symbol is shown for random-ATGC and $\lambda$-DNA,
795: respectively.
796: %
797: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
798: top panel.}
799: \end{figure}
800: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
801: Upon further increasing the disorder to $W=2$, as shown in Fig.\
802: \ref{fig-LM-uniW2-loc_energy}, the localisation lengths continue to
803: decrease. Note that we observe a slight broadening of the bands and
804: states begin to shift into the gap.
805: %
806: We also see that the behaviour of random-ATGC and $\lambda$-DNA is quite
807: similar and at these disorder strengths, even telomeric DNA follows the
808: same trends.
809: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
810: \begin{figure}
811: \centering
812: % \includegraphics[width=\figwidth]{all_seq_BB2_uniform.eps}
813: \includegraphics[width=\figwidth]{fig/fig-LM-uniW2-loc-DOS_energy.eps}
814: \caption{\label{fig-LM-uniW2-loc_energy}
815: Top: $\xi(E)$ as in Fig.\ \ref{fig-LM-uniW1-loc_energy} but with
816: $W=2$. Only every 2nd and 5th symbol is shown for random-ATGC and
817: $\lambda$-DNA, respectively.
818: %
819: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
820: top panel.}
821: \end{figure}
822: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
823: At $W=5$, the localisation lengths have been reduced to a few base-pair
824: separation distances and the differences between all $4$ sequences are
825: very small. The gap has been nearly completely filled as shown by the
826: DOS, albeit with states which have a very small localisation length.
827: This will become important later.
828:
829: Thus, in summary, we have seen that adding uniform disorder onto the
830: backbone leads to a reduction of the localisation lengths and
831: consequently a reduction of the electron conductance. Strictly speaking,
832: all 4 strands are insulators. However, their localisation lengths can
833: remain quite large, larger than in many of the experiments. Thus even
834: the localised electron can contribute towards a finite conductivity for
835: these short sequences. In agreement with experiments, poly(dG)-poly(dC)
836: DNA is the most prominent candidate.
837: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
838: \begin{figure}
839: \centering
840: % \includegraphics[width=\figwidth]{all_seq_BB5_uniform.eps}
841: \includegraphics[width=\figwidth]{fig/fig-LM-uniW5-loc-DOS_energy.eps}
842: \caption{\label{fig-LM-uniW5-loc_energy}
843: Top: $\xi(E)$ as in Fig.\ \ref{fig-LM-uniW1-loc_energy} but with
844: $W=5$. Only every 2nd and 5th symbol is shown for random-ATGC and
845: $\lambda$-DNA, respectively.
846: %
847: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
848: top panel.}
849: \end{figure}
850: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
851:
852: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
853: \subsection{DNA in an ionic solution}
854: \label{sec-binary_energy}
855:
856: When in solution, the negatively charged oxygen on the backbone will
857: attract cations such as Na$^{+}$. This will give rise to a dramatic
858: change in local electronic properties at the oxygen-carrying backbone
859: site, but not necessarily influence the neighbouring sites. The effects
860: at each such site will be the same and thus in contrast to a uniform
861: disorder used in section \ref{sec-uniform_energy}, a binary distribution
862: such as $\varepsilon_{i,q}= \pm W/2$ is more appropriate.
863: %
864: For simplicity, we choose $50\%$ of all backbone sites to be occupied
865: $\varepsilon_{i,q}=-W/2$ while the other half remains empty with
866: $\varepsilon_{i,q}=+W/2$. We note that a mixture of concentrations has
867: been studied in the context of the Anderson model in Ref.\
868: \cite{PlyRS03}.
869:
870: In Fig.\ \ref{fig-LM-binW1-loc_energy}, we show the results for moderate
871: binary disorder. In comparison with the uniformly disordered case of
872: Fig.\ \ref{fig-LM-uniW1-loc_energy}, we see that the localisation
873: lengths have decreased further. This is expected because binary disorder
874: is known to be very strong \cite{PlyRS03}. Also, the gap has already
875: started to fill.
876: %
877: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
878: \begin{figure}
879: \centering
880: % \includegraphics[width=\figwidth]{all_seq_BB1_binary.eps}
881: \includegraphics[width=\figwidth]{fig/fig-LM-binW1-loc-DOS_energy.eps}
882: \caption{\label{fig-LM-binW1-loc_energy}
883: Top: Energy dependence of the localisation lengths, $\xi(E)$, for
884: poly(dG)-poly(dC), telomeric, random-ATGC and $\lambda$-DNA in the
885: presence of {\em binary} backbone disorder with $W=1$. Only every
886: 2nd and 5th symbol is shown for random-ATGC and $\lambda$-DNA,
887: respectively.
888: %
889: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
890: top panel.}
891: \end{figure}
892: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
893:
894: Increasing the disorder leads again to a decrease of $\xi$ in the energy
895: regions corresponding to the bands. Directly at $E=\pm W/2$, we observe
896: $2$ strong peaks in the DOS which is accompanied by reduced localization
897: lengths. This peak corresponds to the infinite potential barrier or well
898: at $E=-W/2$ or $+W/2$, respectively, as indicated by Eq.\
899: (\ref{eq-ladder-effective}). In Fig.\ \ref{fig-LM-binW1-loc_energy},
900: these peaks were not yet visible. We also see in Fig.\
901: \ref{fig-LM-binW2-loc_energy} that the localisation lengths for states
902: in the band centre start to increase to values $\gtrsim 1$.
903: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
904: \begin{figure}
905: \centering
906: % \includegraphics[width=\figwidth]{all_seq_BB2_binary.eps}
907: \includegraphics[width=\figwidth]{fig/fig-LM-binW2-loc-DOS_energy.eps}
908: \caption{\label{fig-LM-binW2-loc_energy}
909: Top: $\xi(E)$ as in Fig.\ \ref{fig-LM-binW1-loc_energy} but with
910: $W=2$. Only every 2nd and 5th symbol is shown for random-ATGC and
911: $\lambda$-DNA, respectively.
912: %
913: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
914: top panel.}
915: \end{figure}
916: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
917: This trend continues for larger $W$ as shown in Fig.\
918: \ref{fig-LM-binW5-loc_energy}. We see a crossover into a regime where
919: the two original, weak-disorder bands have nearly vanished and states in the
920: centre at $E=0$ are starting to show an increasing localisation length
921: {\em upon increasing the binary disorder}.
922: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
923: \begin{figure}
924: \centering
925: % \includegraphics[width=\figwidth]{all_seq_BB5_binary.eps}
926: \includegraphics[width=\figwidth]{fig/fig-LM-binW5-loc-DOS_energy.eps}
927: \caption{\label{fig-LM-binW5-loc_energy}
928: Top: $\xi(E)$ as in Fig.\ \ref{fig-LM-binW1-loc_energy} but with
929: $W=5$. Only every 2nd and 5th symbol is shown for random-ATGC and
930: $\lambda$-DNA, respectively.
931: %
932: Bottom: DOS for $\lambda$-DNA using the same parameters as in the
933: top panel.}
934: \end{figure}
935: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
936: A further increase in $W$ eventually leads to the complete destruction
937: of the original bands and the formation of a single band symmetric
938: around $E=0$ at about $W\sim 2.5$.
939:
940: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
941: \subsection{Delocalisation due to disorder}
942: \label{sec-delocalization}
943:
944: The results of the previous section suggest that increasing the disorder
945: in different regions of the energy will lead to different transport
946: behaviour. Of particular interest is the region at $E=0$. In Fig.\
947: \ref{fig-LM-binE0-loc_disorder} the variation of $\xi$ as a function of
948: binary disorder strength for all different sequences is shown. While
949: $\xi < 1$ for small disorder, we see that upon increasing the disorder,
950: states begin to appear and their localisation lengths increase for all
951: DNA sequences. Thus we indeed observe a counter-intuitive {\em
952: delocalisation} by disorder at $E=0$. As before, poly(dG)-poly(dC) and
953: telomeric disorder show the largest localisation lengths, whereas
954: random-ATGC and $\lambda$-DNA give rise to a smaller and nearly
955: identical effect.
956: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
957: \begin{figure}
958: \centering
959: % \includegraphics[width=\figwidth]{binary_at0_2dmod.eps}
960: \includegraphics[width=\figwidth]{fig/fig-LM-binE0-loc-DOS_disorder.eps}
961: \caption{\label{fig-LM-binE0-loc_disorder}
962: Disorder dependence of $\xi$ for poly(dG)-poly(dC), telomeric,
963: random-ATGC and $\lambda$-DNA at $E=0$. Only every 10th symbol is
964: shown for all sequences. The shaded curve is the corresponding unnormalized DOS
965: for $\lambda$-DNA.}
966: \end{figure}
967: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
968: In Fig.\ \ref{fig-LM-binE3-loc_disorder} we show that this effect does
969: not exist at $E=3$, i.e.\ for energies corresponding to the formerly
970: largest localisation lengths. Rather, at $E=3$, the localisation lengths
971: for all DNA sequences quickly drop to $\xi \sim 1$.
972: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
973: \begin{figure}
974: \centering
975: % \includegraphics[width=\figwidth]{binary_at3_2d.eps}
976: \includegraphics[width=\figwidth]{fig/fig-LM-binE3-loc-DOS_disorder.eps}
977: \caption{\label{fig-LM-binE3-loc_disorder}
978: $\xi(W)$ as in Fig.\ \ref{fig-LM-binE0-loc_disorder} but with $E=3$.
979: Only every 10th symbol is shown for all DNA sequences. The shaded
980: curve is the corresponding unnormalized DOS for $\lambda$-DNA.}
981: \end{figure}
982: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
983: The delocalisation effect is also observed for uniform disorder, but is
984: much smaller. As shown in Fig.\ \ref{fig-FM-uniE0-loc_disorder}, the
985: enhancement is up to about $\xi=1$ for the fishbone model
986: (\ref{eq-fishbone}). Results for the ladder model (\ref{eq-ladder}) are
987: about $1.7$ times larger.
988: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
989: \begin{figure}
990: \centering
991: % \includegraphics[width=\figwidth]{uniform_2d.eps}
992: \includegraphics[width=\figwidth]{fig/fig-FM-uniE0-loc-DOS_disorder.eps}
993: \caption{\label{fig-FM-uniE0-loc_disorder}
994: $\xi(W)$ as in Fig.\ \ref{fig-LM-binE0-loc_disorder} but with
995: uniform disorder at $E=0$ and for the {\em fishbone model}. Only
996: every 10th symbol is shown for all DNA sequences. The shaded curve
997: is the corresponding unnormalized DOS for $\lambda$-DNA.}
998: \end{figure}
999: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1000:
1001: This surprising delocalisation-by-disorder behaviour can be understood
1002: by considering the effects of disorder at the backbone for the effective
1003: Hamiltonians (\ref{eq-fishbone-effective}) and
1004: (\ref{eq-ladder-effective}). At $E=0$, the onsite potential correction
1005: term ${\left(t_{i}^{q}\right)^{2}}/{(\varepsilon_{i}^{q} - E)}$ will
1006: {\em decrease} upon increasing the $\varepsilon_{i}^{q}$ values. For
1007: binary disorders $\varepsilon_{i}^{q} = \pm W/2$, this holds for
1008: $|\varepsilon_{i}^{q}| > |E|$ as shown in Fig.\
1009: \ref{fig-LM-binE3-loc_disorder}. However, for large $|E|$, the
1010: localisation lengths decrease quickly due to the much smaller density of
1011: states. Thus the net effect is an eventual decrease (or an only very
1012: small increase) of $\xi$ for large $E$. Note the dip at
1013: $|\varepsilon_{i}^{q}|=E=3$ in the figure, which corresponds to the
1014: effective $\varepsilon_{i}= \infty$, i.e.\ an infinitely strong trap
1015: yielding extremely strong localisation.
1016: %
1017: For uniform disorder $\varepsilon_{i}^{q} \in [-W/2,W/2]$ --- and
1018: generally any disorder with compact support around $E=0$ --- the above
1019: inequality is never full-filled and even for $E=0$ we will find small
1020: $\varepsilon_{i}^{q} \sim 0$ such that we have strong trapping and
1021: localisation.
1022:
1023: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1024: \section{Investigating the local properties of the sequences}
1025: \label{sec-local}
1026: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1027:
1028: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1029: \subsection{Variation of $\xi$ along the DNA strand}
1030: %\label{sec-variation}
1031:
1032: In the preceding sections, we had computed estimates of the localisation
1033: length $\xi$ for complete DNA strands, i.e.\ the $\xi$ values are {\em
1034: averages}. However, the biological function of DNA clearly depends on
1035: the local structure of the sequence in a paramount way. After all, only
1036: certain parts of DNA code for proteins, while others do not. In
1037: addition, the exact sequence of the bases specifies the protein that is
1038: to be assembled.
1039: %
1040: Thus, in order to gain access to the local properties, we have performed
1041: computations of $\xi$ on subsequences of complete DNA strands. We start
1042: by artificially restricting ourselves to finite windows of length $K=
1043: 10, 30, 50, 100, 200, 500, 1000$ and compute the localisation lengths
1044: $\xi_{K}(r)$ where $r=1, 2, \ldots, L-K$ denotes the starting position
1045: of the window of length $K$.
1046:
1047: In order to see how the exact sequence determines our results, we have
1048: also randomly permuted (scrambled) the $\lambda$-DNA sequence so that
1049: the content of A, T, G, and C bases is the same, but their order is
1050: randomised. Differences in the localisation properties should then
1051: indicate the importance of the exact order.
1052: %
1053: From the biological information available on bacteriophage
1054: $\lambda$-DNA, we compute the localisation length for the coding regions
1055: \cite{DanSSS83} and then for window lengths $K$ that correspond exactly
1056: to the length of each coding region. Again, if the electronic properties
1057: --- as measured by the localisation length --- are linked to biological
1058: content, we would expect to see characteristic differences.
1059:
1060: In Figs.\ \ref{fig-LM-cleanE3-loc_window-100} and
1061: \ref{fig-LM-cleanE3-loc_window-1000}, we show results for $K=100$ and
1062: $1000$, respectively.
1063: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1064: \begin{figure}
1065: \centering
1066: % \includegraphics[width=\figwidth]{lambda0_allwl.eps}
1067: % \includegraphics[width=\figwidth]{fig/fig-LM-cleanE3-loc_window-100.eps}
1068: \includegraphics[width=\figwidth]{fig/fig-LM-cleanE3-loc_window-100.ps}
1069: \caption{\label{fig-LM-cleanE3-loc_window-100}
1070: Top: Variation of the localisation lengths for a sliding window of
1071: length $K=100$ as a function of window starting position for
1072: $\lambda$-DNA at $E=3$. The black crosses ($\times$) denote results
1073: for windows corresponding to the coding sequences of $\lambda$-DNA
1074: only. The dashed horizontal line denotes $K$.
1075: %
1076: Middle: Same as in the top panel but with randomly scrambled
1077: $\lambda$-DNA.
1078: %
1079: Bottom: Normalised distribution functions $P(\xi)$ for the
1080: localisation lengths $\xi$ of $\lambda$- (black) and
1081: scrambled-$\lambda$-DNA (grey). }
1082: \end{figure}
1083: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1084: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1085: \begin{figure}
1086: \centering
1087: % \includegraphics[width=\figwidth]{lambda0_allwl.eps}
1088: % \includegraphics[width=\figwidth]{fig/fig-LM-cleanE3-loc_window-1000.eps}
1089: \includegraphics[width=\figwidth]{fig/fig-LM-cleanE3-loc_window-1000.ps}
1090: \caption{\label{fig-LM-cleanE3-loc_window-1000}
1091: Variation of the localisation lengths for a sliding window of length
1092: $K=1000$ at $E=3$ as in Fig.\
1093: \protect\ref{fig-LM-cleanE3-loc_window-100}.
1094: %
1095: Middle: Same as in the top panel but with randomly scrambled
1096: $\lambda$-DNA.
1097: %
1098: Bottom: Normalised distribution functions $P(\xi)$ for the
1099: localisation lengths $\xi$ of $\lambda$- (black) and
1100: scrambled-$\lambda$-DNA (grey). }
1101: \end{figure}
1102: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1103: From Fig.\ \ref{fig-LM-cleanE3-loc_window-100}, we see from $P(\xi)$
1104: that the localisation lengths for $\lambda$-DNA are mostly distributed
1105: around $15$--$20$, but $P(\xi)$ has a rather long tail for large $\xi$.
1106: However, there are some windows where the localisation lengths exceed
1107: even the size of the window $K=100$. Thus at specific positions in the
1108: DNA sequence, the system appears essentially extended with $\xi > K$. On
1109: the other hand, the distribution $P(\xi)$ is identical when instead of
1110: $\lambda$-DNA, we consider scrambled DNA. Therefore the presence of
1111: such regions is not unique to $\lambda$-DNA. The results from windows
1112: positioned at the coding part of $\lambda$-DNA appear statistically
1113: similar to the complete sequence, i.e.\ including also the non-coding
1114: regions. This suggests that with respect to the localisation properties
1115: there is no obvious difference between $\lambda$-DNA and scrambled
1116: $\lambda$-DNA as well as coding and non-coding regions. We emphasise
1117: that similar results have been obtained for a DNA sequence constructed
1118: from the SARS corona-viral data.
1119:
1120: In Fig.\ \ref{fig-LM-cleanE3-loc_window-100}, we repeat these
1121: calculations but with $K=1000$. Clearly, $P(\xi)$ is peaked again around
1122: $15$--$20$ and this time has no tail. In all cases, $K>\xi$. Again, the
1123: results for scrambled DNA are different in each window, and now even
1124: $P(\xi)$ is somewhat shifted with respect to $\lambda$-DNA.
1125:
1126: Thus in conclusion, we do not see significant differences between
1127: $\lambda$-DNA and its scrambled counter part. Moreover, there appears to
1128: be no large difference between the localisation lengths measured in the
1129: coding and the non-coding sequences of bacteriophage $\lambda$-DNA. This
1130: indicates that the average $\xi$ values computed in the previous
1131: sections is sufficient when considering the electronic localisation
1132: properties of the $4$ complete DNA sequences.
1133:
1134: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1135: \subsection{Computing correlation functions}
1136: %\label{sec-correlation}
1137:
1138: As shown in the last section, the spatial variation of $\xi$ for a fixed
1139: window size is characteristic of the order of bases in the DNA sequence.
1140: Thus we can now study how this biological information is retained at the
1141: level of localisation lengths. In order to do so, we define the
1142: correlation function
1143: \begin{equation}
1144: {\rm Cor}(k)%= \langle \xi(r_i) \xi(r_{i+k}) \rangle
1145: =
1146: \frac{\sum_{i=1}^{n-k}\left[\xi(r_i)-\langle{\xi}\rangle\right]
1147: \left[\xi(r_{i+k})-\langle{\xi}\rangle\right]}%
1148: {\sum_{i=1}^{n}\left[\xi(r_i)-\langle{\xi}\rangle\right]^2}
1149: \label{eq-cor}
1150: \end{equation}
1151: where $\langle{\xi}\rangle={\sum_{i=1}^{n}\xi(r_i)}/{n}$ is $\xi$
1152: averaged over all $n=L-(K-1)$ windows for each of which the individual
1153: localisation lengths are $\xi(r_i)$.
1154:
1155: In Fig.\ \ref{fig-LM-Win-E0-cor_pos} we show the results obtained for
1156: $\lambda$-DNA with windows of length $10$, $100$ and $1000$. We first
1157: note that ${\rm Cor}(k)$ drops rapidly until the distance $k$ exceeds
1158: the window width $K$ (see the inset of Fig.\
1159: \ref{fig-LM-Win-E0-cor_pos}). For $k>K$, ${\rm Cor}(k)$ fluctuates
1160: typically between $\pm 0.2$ and there is a larger anti-correlation for
1161: base-pair separations of about $k=8000$. We note that such large scale
1162: features are not present when considering scrambled $\lambda$-DNA
1163: instead.
1164:
1165: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1166: \begin{figure}
1167: \centering
1168: % \includegraphics[width=\figwidth]{lambda0_allwl.eps}
1169: \includegraphics[width=\figwidth]{fig/fig-LM-Win-E0-cor_pos.eps}
1170: \caption{\label{fig-LM-Win-E0-cor_pos}
1171: ${\rm Cor}(k)$ as defined in Eq.\ (\ref{eq-cor}) for $\lambda$-DNA and
1172: $K=10$, $100$, and $1000$ at $E=0$. The inset shows the same date
1173: but plotted as a function of normalized seperation $k/K$.}
1174: \end{figure}
1175: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1176:
1177: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1178: %\begin{figure}
1179: % \centering
1180: % \includegraphics[width=\figwidth]{scrambledlam0_allwl.eps}
1181: % \caption{scrambled lambda dna all window sizes, ladder.}
1182: %\end{figure}
1183: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1184:
1185: %\begin{figure}
1186: % \centering
1187: % \includegraphics[width=\figwidth]{lambda_scr_comp_30.eps}
1188: % \caption{lambda and scrambled compared wl=30, ladder.}
1189: %\end{figure}
1190:
1191: %\begin{figure}
1192: % \centering
1193: % \includegraphics[width=\figwidth]{lambda_scr_comp_100.eps}
1194: % \caption{lambda and scrambled compared wl=100, ladder.}
1195: %\end{figure}
1196:
1197: %\begin{figure}
1198: % \centering
1199: % \includegraphics[width=\figwidth]{lambda_scr_comp_500.eps}
1200: % \caption{lambda and scrambled compared wl=500, ladder.}
1201: %\end{figure}
1202:
1203: %\begin{figure}
1204: % \centering
1205: % \includegraphics[width=\figwidth]{lambda_scr_comp_1000.eps}
1206: % \caption{lambda and scrambled compared wl=1000, ladder.}
1207: %\end{figure}
1208:
1209: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1210: \section{Discussion}
1211: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1212:
1213: The fishbone and ladder models studied in the present paper give
1214: qualitatively similar results, i.e.\ a gap in the DOS on the order of
1215: the hopping energies to the backbone, extended states for periodic DNA
1216: sequences and localised states for any non-zero disorder strength. Thus
1217: at $T=0$, our results suggest that DNA is an insulator unless perfectly
1218: ordered. Quantitatively, the localisation lengths $\xi$ computed for
1219: the ladder model are larger than for the fishbone model. Since we are
1220: interested in these non-universal lengths, the ladder model is clearly
1221: the more appropriate model.
1222:
1223: % fishbone versus ladder, both gaps, but non-universal lambdas
1224:
1225: The localisation lengths measure the spatial extent of a conducting
1226: electron. Our results suggest --- in agreement with all previous
1227: considerations --- that poly(dG)-poly(dC) DNA allows the largest values
1228: of $\xi$. Even after adding a substantial amount of disorder,
1229: poly(dG)-poly(dC) DNA can still support localization lengths of a few
1230: hundred base-pair seperation lengths. With nanoscopic experiments
1231: currently probing at the most a few dozen bases, this suggests that
1232: poly(dG)-poly(dC) DNA will appear to be conducting in these experiments.
1233:
1234: Furthermore, telomeric DNA is a very encouraging and interesting
1235: naturally occuring sequence because it gives very large localisation
1236: lengths in the weakly disordered regime. Nevertheless, we find that all
1237: investigated, non-periodic DNA sequences such as, e.g.\ random-ATGC and
1238: $\lambda$-DNA, give localised behaviour even in the clean state. This
1239: indicates that they are insulating at $T=0$.
1240:
1241: % disorder effects:
1242: % - reducing the localization lengths
1243: % - still larger than in experiments
1244: % - damage repair enhancement?
1245:
1246: When the effects of the environment, modelled by their potential changes
1247: on the backbone, are included, we find that the localisation lengths in
1248: the two bands decrease quickly upon increasing the disorder.
1249: Nevertheless, depending on the value of the Fermi energy, the resulting
1250: $\xi$ values can still be 10-20 base-pairs long. While this may not
1251: give metallic behavior, it can still result in a finite current for
1252: small sequences. We also note that these distances are quite close to
1253: those obtained from electron-transfer studies.
1254:
1255: % delocalization by disorder at E=0
1256:
1257: The backbone disorder also leads to states moving into the gap.
1258: Therefore the environment prepared in the experiments determines the gap
1259: which is being measured. Furthermore, the localisation properties of the
1260: states in the former gap are drastically different from those in the 2
1261: bands. Increasing the disorder leads to an increase in the localization
1262: lengths and thus potentially larger currents. This is most pronounced
1263: for binary disorder, taken to model the adhesion of cations in solution.
1264: Thus within the $2$ models studied, we find that their transport
1265: properties are in a very crucial way determined by the environment.
1266: Differences in experimental set-up such as measurements in 2D surfaces
1267: or between elevated contacts are likely to lead to quite different
1268: results.
1269:
1270: % sequence dependence
1271: %
1272:
1273: As far as the correlations within biological $\lambda$-DNA are
1274: concerned, we see only a negligible difference between the localisation
1275: properties of the coding and non-coding parts. However, this is clearly
1276: dependent on the chosen energy and the particular window lengths used.
1277: Investigations on other DNA sequences are in progress.
1278:
1279:
1280: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1281: \acknowledgments
1282: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1283: It is a pleasure to thank H.\ Burgert, D.\ Hodgson, M.\ Pfeiffer and D.\
1284: Porath for stimulating discussions.
1285:
1286: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1287: % References
1288: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1289: %\bibliographystyle{prsty}\bibliography{bibliograph}
1290: %\bibliographystyle{biophysj}\bibliography{bibliograph}
1291:
1292: \begin{thebibliography}{60}
1293: \providecommand{\natexlab}[1]{#1}
1294:
1295: \bibitem[{Spartan()}]{Spartan}
1296: \newblock SPARTAN version 5.0, User's Guide.
1297: \newblock Wavefunction Inc., 18401 Von Karman Ave., Suite 370 Irvine, CA 92612.
1298:
1299: \bibitem[{Alberts et~al.(1994)Alberts, Bray, Lewis, Raff, Roberts, and
1300: Watson}]{AlbBLR94}
1301: Alberts, B., D.~Bray, J.~Lewis, M.~Raff, K.~Roberts, and J.~Watson. 1994.
1302: \newblock Molecular Biology of the Cell. Garland, New York.
1303:
1304: \bibitem[{Asabaeva and Tang(2000)}]{AsaT00}
1305: Asabaeva, A., and M.~Tang. 2000.
1306: \newblock Electrical conductivity in oriented {DNA}.
1307: \newblock \emph{National Nanofabrication Users Network Newsletter}. :56--57.
1308:
1309: \bibitem[{Barnett et~al.(2001)Barnett, Cleveland, Joy, Landman, and
1310: Schuster}]{BarCJL01}
1311: Barnett, R.~N., C.~L. Cleveland, A.~Joy, U.~Landman, and G.~B. Schuster. 2001.
1312: \newblock Charge migration in {DNA}: ion-gated transport.
1313: \newblock \emph{Science}. 294:567--571.
1314:
1315: \bibitem[{Berlin et~al.(2000)Berlin, Burin, and Ratner}]{BerBR00}
1316: Berlin, Y.~A., A.~L. Burin, and M.~A. Ratner. 2000.
1317: \newblock On the long-range charge transfer in {DNA}.
1318: \newblock \emph{The Journal of Physical Chemistry}. 104:443--445.
1319:
1320: \bibitem[{Bhalla et~al.(2003)Bhalla, Bajpai, and Bharadwaj}]{BhaBB03}
1321: Bhalla, V., R.~P. Bajpai, and L.~M. Bharadwaj. 2003.
1322: \newblock {DNA} electronics.
1323: \newblock \emph{European Molecular Biology reports}. 4:442--445.
1324:
1325: \bibitem[{Biswas et~al.(2000)Biswas, Cain, {R\"{o}mer}, and
1326: Schreiber}]{BisCRS00}
1327: Biswas, P., P.~Cain, R.~A. {R\"{o}mer}, and M.~Schreiber. 2000.
1328: \newblock Off-diagonal disorder in the {Anderson} model of localization.
1329: \newblock \emph{phys. stat. sol. (b)}. 218:205--209.
1330: \newblock {ArXiv}: cond-mat/0001315.
1331:
1332: \bibitem[{Bixon et~al.(1999)Bixon, Giese, Wessely, Langenbacher,
1333: Michel-Beyerle, and Jortner}]{BixGWL99}
1334: Bixon, M., B.~Giese, S.~Wessely, T.~Langenbacher, M.~E. Michel-Beyerle, and
1335: J.~Jortner. 1999.
1336: \newblock Long-range charge hopping in {DNA}.
1337: \newblock \emph{PNAS}. 96:11713.
1338:
1339: \bibitem[{Boon et~al.(2003)Boon, Livingston, Chmiel, David, and
1340: Barton}]{BooLCD03}
1341: Boon, E., A.~Livingston, N.~Chmiel, S.~David, and J.~Barton. 2003.
1342: \newblock {DNA}-mediated charge transport for {DNA} repair.
1343: \newblock \emph{Proc. Nat. Acad. Sci.} 100:12543--12547.
1344:
1345: \bibitem[{Braun et~al.(1998)Braun, Eichen, Sivan, and Ben-Yoseph}]{BraESB98}
1346: Braun, E., Y.~Eichen, U.~Sivan, and G.~Ben-Yoseph. 1998.
1347: \newblock {DNA}-templated assembly and electrode attachment of a conducting
1348: silver wire.
1349: \newblock \emph{Nature}. 391:775--778.
1350:
1351: \bibitem[{Bruinsma et~al.(2000)Bruinsma, Gruner, D'Orsogna, and
1352: Rudnick}]{BruGOR00}
1353: Bruinsma, R., G.~Gruner, M.~R. D'Orsogna, and J.~Rudnick. 2000.
1354: \newblock Fluctuation-facilitated charge migration along {DNA}.
1355: \newblock \emph{Phys. Rev. Lett.} 85:4393--4396.
1356:
1357: \bibitem[{Cuenda and Sanchez(2004)}]{CueS04}
1358: Cuenda, S., and A.~Sanchez. 2004.
1359: \newblock Disorder and fluctuations in nonlinear excitations in {DNA}. {ArXiv}:
1360: q-bio.BM/0403003v1.
1361:
1362: \bibitem[{Cuniberti et~al.(2002)Cuniberti, Craco, Porath, and
1363: Dekker}]{CunCPD02}
1364: Cuniberti, G., L.~Craco, D.~Porath, and C.~Dekker. 2002.
1365: \newblock Backbone-induced semiconducting behavior in short {DNA} wires.
1366: \newblock \emph{Phys. Rev. B}. 65:241314--4.
1367:
1368: \bibitem[{Daniels et~al.(1983)Daniels, Schroeder, Szybalski, Sanger, and
1369: Blattner}]{DanSSS83}
1370: Daniels, D.~L., J.~L. Schroeder, W.~Szybalski, F.~Sanger, and F.~R. Blattner.
1371: 1983.
1372: \newblock LAMBDA II:469-517, chapter Appendix I: A molecular map of colphase
1373: lambda.
1374: \newblock Cold Spring Harbor Laboratory, Cold Spring Harbor.
1375:
1376: \bibitem[{Davies and Inglesfield(2004)}]{DavI04}
1377: Davies, O.~R., and J.~E. Inglesfield. 2004.
1378: \newblock Embedding methods for conductance in {DNA}.
1379: \newblock \emph{Phys. Rev. B}. 69:195110--13.
1380:
1381: \bibitem[{Dekker and Ratner(2001)}]{DekR01}
1382: Dekker, C., and M.~A. Ratner. 2001.
1383: \newblock Electronic properties of {DNA}.
1384: \newblock \emph{Physics World}. 14:29.
1385:
1386: \bibitem[{Delaney and Barton(2003)}]{DelB03}
1387: Delaney, S., and J.~K. Barton. 2003.
1388: \newblock Long-range {DNA} charge transport.
1389: \newblock \emph{J. Org. Chem.} 68:6475--6483.
1390:
1391: \bibitem[{Dyson(1953)}]{Dys53}
1392: Dyson, F.~J. 1953.
1393: \newblock The dynamics of a disordered linear chain.
1394: \newblock \emph{Phys. Rev.} 92:1331--1338.
1395:
1396: \bibitem[{Eilmes et~al.(1998)Eilmes, {{R\"{o}mer}}, and Schreiber}]{EilRS98a}
1397: Eilmes, A., R.~A. {{R\"{o}mer}}, and M.~Schreiber. 1998.
1398: \newblock The two-dimensional {Anderson} model of localization with random
1399: hopping.
1400: \newblock \emph{Eur. Phys. J. B}. 1:29--38.
1401:
1402: \bibitem[{Endres et~al.(2004)Endres, Cox, and Singh}]{EndCS04}
1403: Endres, R.~G., D.~L. Cox, and R.~P. Singh. 2004.
1404: \newblock The quest for high-conductance {DNA}.
1405: \newblock \emph{Rev. Mod. Phys.} 76:195--214.
1406:
1407: \bibitem[{Fink and C.(1999)}]{FinS99}
1408: Fink, H.-W., and S.~C. 1999.
1409: \newblock Electrical conduction through {DNA} molecules.
1410: \newblock \emph{Nature}. 398:407--410.
1411:
1412: \bibitem[{Frahm et~al.(1995)Frahm, {M\"{u}ller-Groeling}, Pichard, and
1413: Weinmann}]{FraMPW95}
1414: Frahm, K., A.~{M\"{u}ller-Groeling}, J.~L. Pichard, and D.~Weinmann. 1995.
1415: \newblock Scaling in interaction-assisted coherent transport.
1416: \newblock \emph{Europhys. Lett.} 31:169.
1417:
1418: \bibitem[{Garzon et~al.(2001)Garzon, Artacho, Beltran, Garcia, Junquera,
1419: Michaelian, Ordejon, Rovira, Sanchez-Portal, and Soler}]{GarABG01}
1420: Garzon, I.~L., E.~Artacho, M.~R. Beltran, A.~Garcia, J.~Junquera,
1421: K.~Michaelian, P.~Ordejon, C.~Rovira, D.~Sanchez-Portal, and J.~M. Soler.
1422: 2001.
1423: \newblock Hybrid {DNA}-gold nanostructured materials: an ab-initio approach.
1424: \newblock \emph{Nanotechnology}. 12:126--131.
1425:
1426: \bibitem[{Gutierrez et~al.(2004)Gutierrez, Mandal, and Cuniberti}]{GutMC04}
1427: Gutierrez, R., S.~Mandal, and G.~Cuniberti. 2004.
1428: \newblock Quantum transport in {DNA} wires: Influence of a strong dissipative
1429: environment. {ArXiv}: cond-mat/0410660.
1430:
1431: \bibitem[{Kelley and Barton(1999)}]{KelB99}
1432: Kelley, S.~O., and J.~K. Barton. 1999.
1433: \newblock Electron transfer between bases in double helical {DNA}.
1434: \newblock \emph{Science}. 283:375--381.
1435:
1436: \bibitem[{Kramer and MacKinnon(1993)}]{KraM93}
1437: Kramer, B., and A.~MacKinnon. 1993.
1438: \newblock Localization: theory and experiment.
1439: \newblock \emph{Rep. Prog. Phys.} 56:1469--1564.
1440:
1441: \bibitem[{MacKinnon(1980)}]{Mac80}
1442: MacKinnon, A. 1980.
1443: \newblock The conductivity of the one-dimensional disordered anderson model: a
1444: new numerical method.
1445: \newblock \emph{J. Phys.: Condens. Matter}. 13:L1031--L1034.
1446:
1447: \bibitem[{MacKinnon(1985)}]{Mac85}
1448: MacKinnon, A. 1985.
1449: \newblock The calculation of transport properties and density of states of
1450: disordered solids.
1451: \newblock \emph{Z. Phys. B}. 59:385--390.
1452:
1453: \bibitem[{MacKinnon(1994)}]{Mac94}
1454: MacKinnon, A. 1994.
1455: \newblock Critical exponents for the metal-insulator transition.
1456: \newblock \emph{J. Phys.: Condens. Matter}. 6:2511--2518.
1457:
1458: \bibitem[{MacKinnon and Kramer(1983)}]{MacK83}
1459: MacKinnon, A., and B.~Kramer. 1983.
1460: \newblock The scaling theory of electrons in disordered solids: additional
1461: numerical results.
1462: \newblock \emph{Z. Phys. B}. 53:1--13.
1463:
1464: \bibitem[{Murphy et~al.(1993)Murphy, Arkin, Jenkins, Ghatlia, Bossman, Turro,
1465: and Barton}]{MurAJG93}
1466: Murphy, C.~J., M.~A. Arkin, Y.~Jenkins, N.~D. Ghatlia, S.~Bossman, N.~J. Turro,
1467: and J.~K. Barton. 1993.
1468: \newblock Long-range photoinduced electron transfer through a {DNA} helix.
1469: \newblock \emph{Science}. 262:1025--1029.
1470:
1471: \bibitem[{Nakao et~al.(2003)Nakao, Gad, Sugiyama, Otobe, and Ohtani}]{NakGSO03}
1472: Nakao, H., M.~Gad, S.~Sugiyama, K.~Otobe, and T.~Ohtani. 2003.
1473: \newblock Transfer-printing of highly aligned {DNA} nanowires.
1474: \newblock \emph{J. Am. Chem. Soc.} 125:7162--7163.
1475:
1476: \bibitem[{Ndawana et~al.(2004)Ndawana, {R\"{o}mer}, and Schreiber}]{NdaRS04}
1477: Ndawana, M.~L., R.~A. {R\"{o}mer}, and M.~Schreiber. 2004.
1478: \newblock Effects of scale-free disorder on the {Anderson} metal-insulator
1479: transition.
1480: \newblock \emph{Europhys. Lett.} 68:678--684.
1481:
1482: \bibitem[{Okahata et~al.(1998)Okahata, Kobayashi, Tanaka, and
1483: Shimomura}]{OkaKTS98}
1484: Okahata, Y., T.~Kobayashi, K.~Tanaka, and M.~Shimomura. 1998.
1485: \newblock Anisotropic electric conductivity in an aligned {DNA} cast film.
1486: \newblock \emph{J. Am. Chem. Soc.} 120:6165--6166.
1487:
1488: \bibitem[{O'Neil et~al.(2004)O'Neil, Dohno, and Barton}]{OneDB04}
1489: O'Neil, M.~A., C.~Dohno, and J.~K. Barton. 2004.
1490: \newblock Direct chemical evidence for charge transfer between photoexcited
1491: 2-aminopurine and guanine in duplex {DNA}.
1492: \newblock \emph{Journal of the American Chemical Society Communications}.
1493: 126:1316--1317.
1494:
1495: \bibitem[{O'Neil and Fielden(1993)}]{OneF93}
1496: O'Neil, P., and E.~M. Fielden. 1993.
1497: \newblock Primary free radical processes in {DNA}.
1498: \newblock \emph{Adv. Radiat. Biol.} 17:53--120.
1499:
1500: \bibitem[{Pablo et~al.(2000)Pablo, Moreno-Herrero, Colchero, Gomez~Herrero,
1501: Hererro, Baro, Soler, and Artacho}]{PabMCH00}
1502: Pablo, P.~J., F.~Moreno-Herrero, J.~Colchero, J.~Gomez~Herrero, P.~Hererro,
1503: P.~Baro, A. M. an~Ordejon, J.~M. Soler, and E.~Artacho. 2000.
1504: \newblock Absence of dc-conductivity in {$\lambda$-{DNA}}.
1505: \newblock \emph{Phys. Rev. Lett.} 85:4992--4995.
1506:
1507: \bibitem[{Peng et~al.(1992)Peng, Buldyrev, Goldberger, Havlin, Sciortino,
1508: Simons, and Stanley}]{PenBGH92}
1509: Peng, C.~K., S.~Buldyrev, A.~Goldberger, S.~Havlin, F.~Sciortino, M.~Simons,
1510: and H.~E. Stanley. 1992.
1511: \newblock Long-range correlations in nucleotide sequences.
1512: \newblock \emph{Nature}. 356:168--171.
1513:
1514: \bibitem[{Peyrard(2004)}]{Pey04}
1515: Peyrard, M. 2004.
1516: \newblock Nonlinear dynamics and statistical physics of {DNA}.
1517: \newblock \emph{Nonlinearity}. 17:R1--R40.
1518:
1519: \bibitem[{Pichard and Sarma(1981{\natexlab{a}})}]{PicS81a}
1520: Pichard, J.-L., and G.~Sarma. 1981{\natexlab{a}}.
1521: \newblock Finite-size scaling approach to {Anderson} localisation.
1522: \newblock \emph{J. Phys. C}. 14:L127--L132.
1523:
1524: \bibitem[{Pichard and Sarma(1981{\natexlab{b}})}]{PicS81b}
1525: Pichard, J.-L., and G.~Sarma. 1981{\natexlab{b}}.
1526: \newblock Finite-size scaling approach to {Anderson} localisation: {II.}
1527: quantitative analysis and new results.
1528: \newblock \emph{J. Phys. C}. 14:L617--L625.
1529:
1530: \bibitem[{Plyushchay et~al.(2003)Plyushchay, {R\"{o}mer}, and
1531: Schreiber}]{PlyRS03}
1532: Plyushchay, I., R.~A. {R\"{o}mer}, and M.~Schreiber. 2003.
1533: \newblock The three-dimensional anderson model of localization with binary
1534: random potential.
1535: \newblock \emph{Phys. Rev. B}. 68:064201--8.
1536:
1537: \bibitem[{Porath et~al.(2000)Porath, Bezryadin, Vries, and Dekker}]{PorBVD00}
1538: Porath, D., A.~Bezryadin, S.~Vries, and C.~Dekker. 2000.
1539: \newblock Direct measurement of electrical transport through {DNA} molecules.
1540: \newblock \emph{Nature}. 403:635--638.
1541:
1542: \bibitem[{Porath et~al.(2004)Porath, Cuniberti, and Di~Felice}]{PorCD04}
1543: Porath, D., G.~Cuniberti, and R.~Di~Felice. 2004.
1544: \newblock Charge transport in {DNA}-based devices.
1545: \newblock \emph{Topics in Current Chemistry}. 237:183.
1546:
1547: \bibitem[{Rakitin et~al.(2001)Rakitin, Aich, Papadopoulos, Kobzar, Vendeneev,
1548: Lee, and Xu}]{RakAPK01}
1549: Rakitin, A., P.~Aich, C.~Papadopoulos, Y.~Kobzar, A.~Vendeneev, J.~Lee, and
1550: J.~Xu. 2001.
1551: \newblock Metallic conduction through engineered {DNA}: {DNA} nanoelectric
1552: building blocks.
1553: \newblock \emph{Phys. Rev. Lett.} 86:3670--3673.
1554:
1555: \bibitem[{Retel et~al.(1993)Retel, Hoebee, Braun, Lutgerink, Van~der Akker,
1556: Wanamarta, Joenjie, and Lafleur}]{RetHBL93}
1557: Retel, J., B.~Hoebee, J.~E.~F. Braun, J.~T. Lutgerink, E.~Van~der Akker,
1558: H.~Wanamarta, H.~Joenjie, and M.~V.~M. Lafleur. 1993.
1559: \newblock Mutational specificity of oxidative {DNA} damage.
1560: \newblock \emph{Mutation Res.} 299:165.
1561:
1562: \bibitem[{Roche(2003)}]{Roc03}
1563: Roche, S. 2003.
1564: \newblock Sequence dependent {DNA}-mediated conduction.
1565: \newblock \emph{Phys. Rev. Lett.} 91:108101--4.
1566:
1567: \bibitem[{Roche et~al.(2003)Roche, Bicout, {Maci\'{a}}, and Kats}]{RocBMK03}
1568: Roche, S., D.~Bicout, E.~{Maci\'{a}}, and E.~Kats. 2003.
1569: \newblock Long range correlation in {DNA}: Scaling properties and charge
1570: transfer efficiency.
1571: \newblock \emph{Phys. Rev. Lett.} 91:228101--4.
1572:
1573: \bibitem[{{{R\"{o}mer}} and Schreiber(1997)}]{RomS97b}
1574: {{R\"{o}mer}}, R.~A., and M.~Schreiber. 1997.
1575: \newblock The enhancement of the localization length for two-interacting
1576: particles is vanishingly small in transfer-matrix calculations.
1577: \newblock \emph{Phys. Rev. Lett.} 78:4890.
1578:
1579: \bibitem[{{R\"{o}mer} and Schreiber(2003)}]{RomS03}
1580: {R\"{o}mer}, R.~A., and M.~Schreiber. 2003.
1581: \newblock The Anderson Transition and its Ramifications --- Localisation,
1582: Quantum Interference, and Interactions, chapter Numerical investigations of
1583: scaling at the Anderson transition.
1584: \newblock Springer, Berlin, 3--19.
1585:
1586: \bibitem[{{R\"{o}mer} et~al.(2002){R\"{o}mer}, Villagonzalo, and
1587: MacKinnon}]{RomMV03}
1588: {R\"{o}mer}, R.~A., C.~Villagonzalo, and A.~MacKinnon. 2002.
1589: \newblock Thermoelectric properties of disordered systems.
1590: \newblock \emph{J. Phys. Soc. Japan}. 72:167--168.
1591: \newblock Suppl. A.
1592:
1593: \bibitem[{Treadway et~al.(2002)Treadway, Hill, and Barton}]{TreHB01}
1594: Treadway, C.~R., M.~G. Hill, and J.~K. Barton. 2002.
1595: \newblock Charge transport through a molecular $\pi$-stack: Double helical
1596: {DNA}.
1597: \newblock \emph{Chemical Physics}. 281:409--428.
1598:
1599: \bibitem[{Walet and Zakrzewski(2004)}]{WalZ04}
1600: Walet, N.~R., and W.~J. Zakrzewski. 2004.
1601: \newblock A simple model of the charge transfer in {DNA}-like substances.
1602: {ArXiv}: cond-mat/0402059v1.
1603:
1604: \bibitem[{Wan et~al.(2000)Wan, Fiebig, Schiemann, Barton, and
1605: Zewail}]{WanFSB00}
1606: Wan, C., T.~Fiebig, O.~Schiemann, J.~K. Barton, and A.~H. Zewail. 2000.
1607: \newblock Femptosecond direct observation of charge transfer between bases in
1608: {{DNA}}.
1609: \newblock \emph{Proc. Natl. Acad. Sci.} 97:14052--14055.
1610:
1611: \bibitem[{Wang et~al.(2004)Wang, Lewis, and Sankey}]{WanLS04}
1612: Wang, H., J.~P. Lewis, and O.~F. Sankey. 2004.
1613: \newblock Band-gap tunneling states in {DNA}.
1614: \newblock \emph{Phys. Rev. Lett.} 93:016401.
1615:
1616: \bibitem[{Yamada et~al.(2004)Yamada, Starikov, Hennig, and Archilla}]{YamSHA04}
1617: Yamada, H., E.~B. Starikov, D.~Hennig, and J.~F.~R. Archilla. 2004.
1618: \newblock Localization properties of electronic states in polaron model for
1619: poly(dg)-poly(dc) and poly(da)-poly(dt) {DNA} polymers. {ArXiv}:
1620: cond-mat/0407148v1.
1621:
1622: \bibitem[{Yu and Song(2001)}]{YuS01}
1623: Yu, Z., and X.~Song. 2001.
1624: \newblock Variable range hopping and electrical conductivity along the {DNA}
1625: double helix.
1626: \newblock \emph{Phys. Rev. Lett.} 86:6018--6021.
1627:
1628: \bibitem[{Zhang and Ulloa(2004{\natexlab{a}})}]{ZhaU04a}
1629: Zhang, W., and S.~E. Ulloa. 2004{\natexlab{a}}.
1630: \newblock Extended states in disordered systems: Role of off-diagonal
1631: correlations.
1632: \newblock \emph{Phys. Rev. B}. 69:153203.
1633:
1634: \bibitem[{Zhang and Ulloa(2004{\natexlab{b}})}]{ZhaU04b}
1635: Zhang, W., and S.~E. Ulloa. 2004{\natexlab{b}}.
1636: \newblock Structural and dynamical disorder and charge transport in {DNA}.
1637: \newblock \emph{Microelectronics Journal}. 35:23--26.
1638:
1639: \bibitem[Zhong(2003)]{Zho03}
1640: Zhong, J. 2003.
1641: \newblock Electronic transport in {DNA} molecules with backbone disorder.
1642: \newblock \emph{Phys. Rev. B}. (private communication).
1643:
1644: \bibitem{perpendiculartolongaxis} We note that Ref.\ \cite{WalZ04} assumes
1645: transport is via the sugar-phosphate backbone.
1646:
1647: \bibitem{fishbone} The results for the fishbone and ladder models are
1648: qualitatively the same. Quantitatively, the ladder model results have a
1649: nearly twice larger localisation length. This factor approaches $2$, if
1650: $t_{1,2}\rightarrow 0$. Therefore we will focus our discussion on the
1651: 2-channel ladder model.
1652:
1653: \bibitem[Bacteriophage lambda]{lambda} Bacteriophage lambda, complete genome
1654: \verb+[gi|9626243|ref|NC_001416.1|[9626243]]+, Genbank Accession number
1655: \verb+NC_001416+, \verb+http://www.ncbi.nlm.nih.gov/entrez/+
1656:
1657: \bibitem[SARS]{sars} SARS coronavirus, complete genome
1658: \verb+[gi|30271926|ref|NC_004718.3|[30271926]]+, Genbank Accession number
1659: \verb+NC_004718+, \verb+http://www.ncbi.nlm.nih.gov/entrez/+
1660:
1661: \bibitem[CEN2]{cen2} CEN2, Chromosome II centromere, \verb+http://www.yeastgenome.org/+
1662:
1663: \end{thebibliography}
1664:
1665:
1666: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1667: \end{document}
1668: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1669:
1670: \begin{thebibliography}{10}
1671: \frenchspacing
1672:
1673: \bibitem{perpendiculartolongaxis} We note that Ref.\ \cite{WalZ04} assumes
1674: transport is via the sugar-phosphate backbone.
1675:
1676: \bibitem{fishbone} The results for the fishbone and ladder models are
1677: qualitatively the same. Quantitatively, the ladder model results have a
1678: nearly twice larger localisation length. This factor approaches $2$, if
1679: $t_{1,2}\rightarrow 0$. Therefore we will focus our discussion on the
1680: 2-channel ladder model.
1681:
1682: \bibitem[Lambda]{lambda} Bacteriophage lambda, complete genome
1683: \verb+[gi|9626243|ref|NC_001416.1|[9626243]]+, Genbank Accession number
1684: \verb+NC_001416+, \verb+http://www.ncbi.nlm.nih.gov/entrez/+
1685:
1686: \bibitem[SARS]{sars} SARS coronavirus, complete genome
1687: \verb+[gi|30271926|ref|NC_004718.3|[30271926]]+, Genbank Accession number
1688: \verb+NC_004718+, \verb+http://www.ncbi.nlm.nih.gov/entrez/+
1689:
1690: \bibitem[CEN2]{cen2} CEN2, Chromosome II centromere, \verb+http://www.yeastgenome.org/+
1691:
1692: %\bibitem[a10]{a10} I am thinking af Bartons group, seem to remember $37$ as a distance.
1693:
1694: \end{thebibliography}
1695: \begin{thebibliography}{10}
1696: \frenchspacing
1697:
1698: \bibitem{FinS99}
1699: H.-W. Fink and C. Schonenberger, Nature {\bf 398}, 407 (1999).
1700:
1701: \bibitem{PorBVD00}
1702: D. Porath, A. Bezryadin, S. Vries, and C. Dekker, Nature {\bf
1703: 403}, 635
1704: (2000).
1705:
1706: \bibitem{ParKSD95}
1707: H. Park, S. Kim, A. Sancar, and J. Deisenhofer, Science {\bf 268},
1708: 1866
1709: (1995).
1710:
1711: \bibitem{OneF93}
1712: P. O'Neil and E.~M. Fielden, Adv. Radiat. Biol. {\bf 17}, 53
1713: (1993).
1714:
1715: \bibitem{RetHBL93}
1716: J. Retel {\it et~al.}, Mutation Res. {\bf 299}, 165 (1993).
1717:
1718: \bibitem{PorCD04}
1719: D. Porath, G. Cuniberti, and R. Di~Felice, Topics in Current
1720: Chemistry {\bf
1721: 237}, 183 (2004).
1722:
1723: \bibitem{DekR01}
1724: C. Dekker and M.~A. Ratner, Physics World {\bf 14}, 29 (2001).
1725:
1726: \bibitem{CunCPD02}
1727: G. Cuniberti, L. Craco, D. Porath, and C. Dekker, Phys. Rev. B
1728: {\bf 65},
1729: 241314 (2002).
1730:
1731: \bibitem{Zho03}
1732: J. Zhong, (private comminucation).
1733:
1734: \bibitem{BruGOR00}
1735: R. Bruinsma, G. Gruner, M.~R. D'Orsogna, and J. Rudnick, Phys.
1736: Rev. B {\bf 85},
1737: 4393 (2000).
1738:
1739: \bibitem{PriRB96}
1740: S. Priyadarshy, S.~M. Risser, and D.~N. Beratan, J. Phys. Chem.
1741: {\bf 100},
1742: 17678 (1996).
1743:
1744: \bibitem{PabMCH00}
1745: P.~J. Pablo {\it et~al.}, Phys. Rev. Lett. {\bf 85}, 4992
1746: (2000).
1747:
1748: \bibitem{Roc03}
1749: S. Roche, Phys. Rev. Lett. {\bf 91}, 108101 (2003);
1750:
1751: \bibitem{1ZhaU04}
1752: W. Zhang and S.~E. Ulloa, Phys. Rev. B {\bf 69}, 153203 (2004).
1753:
1754: \bibitem{2ZhaU04}
1755: W. Zhang and S.~E. Ulloa, Microelectronics Journal {\bf 35}, 23
1756: (2004).
1757:
1758: \bibitem{RocBMK03}
1759: S. Roche, D. Bicout, E. Macia, and E. Kats, Phys. Rev. Lett. {\bf
1760: 91}, 228101
1761: (2003).
1762:
1763: \bibitem{WanLS04}
1764: H. Wang, J.~P. Lewis, and O.~F. Sankey, Phys. Rev. Lett. {\bf 93},
1765: 016401
1766: (2004).
1767:
1768: \bibitem{KraM93}
1769: B. Kramer and A. MacKinnon, Rep. Prog. Phys. {\bf 56}, 1469
1770: (1993).
1771:
1772: \bibitem{YuS01}
1773: Z. Yu and X. Song, Phys. Rev. Lett. {\bf 86}, 6018 (2001).
1774:
1775: \bibitem{BarCJL01}
1776: R.~N. Barnett, C.~L. Cleveland, A. Joy, U. Landman, and G.~B.
1777: Schuster, Science {\bf 294}, 567 (2001).
1778:
1779: \bibitem{BooLCD03}
1780: E. Boon, A. Livingston, N. Chmiel, S. David, and J. Barton, PNAS
1781: {\bf 100}, 12543 (2003).
1782:
1783: \bibitem{BraESB98}
1784: E. Braun, Y. Eichen, U. Sivan, and G. Ben-Yoseph, Nature {\bf
1785: 391}, 775 (1998).
1786:
1787: \bibitem{RakAPK01}
1788: A. Rakitin, P. Aich, C. Papadopoulos, Y. Kobzar, A.~S. Vendeneev,
1789: J.~S. Lee, and J.~M. Xu, PRL {\bf 86}, 3670 (2001).
1790:
1791: \bibitem{NakGSO03}
1792: H. Nakao, M. Gad, S. Sugiyama, K. Otobe, and T. Ohtani, PRL {\bf
1793: 125}, 7162 (2003).
1794:
1795: \bibitem{AsaT00}
1796: A. Asabaeva, and M. Tang, National Nanofabrication, 56 (2000).
1797:
1798: \bibitem{OkaKTS98}
1799: Y. Okahata, T. Kobayashi, K. Tanaka, and M. Shimomura, JACS {\bf
1800: 120}, 6165 (1998).
1801:
1802: \bibitem{BixGWL99}
1803: M. Bixon, B. Giese, S. Wessely, T. Langenbacher, M.~E.
1804: Michel-Beyerle, and J. Jortner, PNAS {\bf 96}, 11713 (1999).
1805:
1806: \bibitem{MurAJG93}
1807: C.~J. Murphy, M.~A. Arkin, Y. Jenkins, N.~D. Ghatlia, S. Bossman,
1808: N.~J. Turro, and J.~K. Barton, Science {\bf 262}, 1025 (1993)
1809:
1810: \bibitem{Rom99}
1811: R. A. {R\"{o}mer}, {Habilitationsschrift}, {Technische}
1812: {Universit\"{a}t} {Chemnitz} (1999).
1813:
1814: \end{thebibliography}
1815:
1816:
1817: