A computational study of bacterial gene regulation and adaptation ...

More documents

Recommendations

Info

Nucleic Acids Research Advance Access published January 18, 2008Nucleic Acids Research, 2008, 1–8doi:10.1093/nar/gkm1181Codon choice in genes depends on flankingsequence information—implications for theoreticalreverse translationKarthikeyan Sivaraman 1 , AswinSaiNarain Seshasayee 2 ,Patrick M. Tarwater 3 and Alexander M. Cole 1, *1 Department of Molecular Biology and Microbiology, Burnett School of Biomedical Sciences, University of CentralFlorida, Orlando, FL, 32816, USA, 2 Genomics and Regulatory Systems Group, EMBL-European BioinformaticsInstitute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK and 3 Department of Biostatistics,University of Texas School of Public Health, El Paso, TX, 79902, USAReceived September 26, 2007; Revised December 7, 2007; Accepted December 27, 2007ABSTRACTAlgorithms for theoretical reverse translation havedirect applications in degenerate PCR. The conventionalpractice is to create several degenerateprimers each of which variably encode the peptideregion of interest. In the current work, for eachcodon we have analyzed the flanking residues inproteins and determined their influence on codonchoice. From this, we created a method for theoreticalreverse translation that includes informationfrom flanking residues of the protein in question.Our method, named the neighbor correlation method(NCM) and its enhancement, the consensus-NCM(c-NCM) performed significantly better than theconventional codon-usage statistic method (CSM).Using the methods NCM and c-NCM, we were ableto increase the average sequence identity from 77%up to 81%. Furthermore, we revealed a significantincrease in coverage, at 80% identity, from _ 20%(CSM) to ` 75% (c-NCM). The algorithms, theirapplications and implications are discussed herein.INTRODUCTIONWord usage and codon usage in bacterial genomes hasbeen extensively documented, both in the coding (1) andnon-coding regions (2). These reports show that wordusage in genomes is non-random and it serves as abiological signature of the organism in question. One suchsignature is codon usage in open reading frames (ORFs),and is reflected in measures such as the codon adaptationindex (CAI) (3). Though CAI provides a convenientmeasure of codon bias, several reports show that codonusage is not a property of isolated codons and in severalcases the bases immediately upstream or downstreamaffect the translation (4). Such neighboring base effects arewell studied in case of stop codon read-through experimentswhere the flanking base or codon has been shown toaffect the accuracy and magnitude of read-through (5).Apart from single bases, the effect of flanking codons hasalso been well studied in literature. Gutman and Hatfield(6) show that there is a strong first-order Markovianrelationship between codons in a gene and this relation isseen even after translation, in proteins. Boycheva andcolleagues extended this study to reveal that translationefficiency is strongly dependent on the dicodon pair thatencodes for a given amino acid pair (7). They suggest thatrelative orientations of t-RNA in the ribosome may causethe observed differences in translation efficiency andsubsequently certain dicodon pairs are selected evolutionarily.Moura and coworkers use a more recent andlarger dataset for an analysis of dicodon usage patterns inboth prokaryotes and eukaryotes. Their results suggestthat the geometric constraints imposed by the translationmachinery are driving forces in the evolution of genesequences in bacteria (8). Collectively, these results suggestthe existence of strong first-order Markovian relationshipsbetween codons in a gene. We hypothesized thatinformation content of such correlations is carried overto the proteins, at least in part, when the gene istranslated. This information manifests itself as a lack ofrandomness in the choice of codons and it is apparentwhen one attempts to theoretically reverse translate aprotein sequence.Reverse translation has been discussed earlier as anabstract logical flow of information from proteins toDNA (9). In this work, we consider the pragmaticproblem of theoretical reverse translation itself, ratherthan that of information flow from proteins to DNA.Theoretical reverse translation of protein sequences haspotential applications in primer design for degenerate*To whom correspondence should be addressed. Tel: 407 823 3633; 407 823 3635; Email: acole@mail.ucf.eduß 2008 The Author(s)This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
2 Nucleic Acids Research, 2008PCR and in design of synthetic genes (10). In degeneratePCR, several primers are designed, each representing avariant DNA sequence encoding the peptide region ofinterest. One of the best methods designed for degeneratePCR can, in the best case scenarios, still utilize up to 128primers on one end (5 0 - or 3 0 -end) and one or more at theother end (11). Though no specific software is availablefor reverse translation, the conventional procedure isto substitute codons for residues based on the overallgenomic codon usage probabilities which required differentprimers be designed for each ambiguous codon in thegene in the region of interest. In practice, it is common foralmost all possibilities to be covered, increasing thenumber of required primers exponentially. Thus, improvementsin reverse translation will help reduce the ambiguityin degenerate PCR.Improvements in reverse translation can be broughtabout by studying the rules of codon usage in the genome,which is feasible due to availability of whole genomesequences. In this study, we created a framework forreverse translation of bacterial gene sequences and term itthe neighbor correlation method (NCM), due to its useof neighboring (flanking) sequence information to predictcodon usage. We provide evidence for the dependency ofcodon choice on the flanking amino acid residues and usedthis dependency to reverse-translate protein sequencesfrom two model genomes. We confirmed that NCM was asubstantial improvement over the conventional method(codon-usage statistic method—CSM). Furthermore,we introduced a modification to both CSM and NCM[consensus CSM (c-CSM) and consensus NCM (c-NCM)]to improve significantly the sensitivity of reverse translationsby both CSM and NCM, and show that theseobserved differences in performance are statisticallysignificant. Finally, using the protein sequences ofSalmonella typhi CT18 and the probability matrix fromEscherichia coli K12, we show that it is possible to reversetranslate sequences from organisms for which a reversetranslation matrix is not available, by using a matrix froma related organism.MATERIALS AND METHODSAll sequences were obtained from the NCBI database. Forthe analyses, the genome and predicted ORF sequences ofE. coli K12 (12), B. subtilis (13), and S. typhi CT18 (14),Acidobacteria bacterium (NC_008095), Aquifex aeolicus(15), Bacteroides thetaiotaomicron (16), Bordetella pertussis(17), Campylobacter jejuni (18), Caulobacter crescentus(19), Chlamydia trachomatis (20), Clostridium acetobutylicum(21), Dehalococcoides ehtenogenes (22), Deinococcusradiodurans (23), Fusobacterium nucleatum (24),Lactobacillus acidophilus (25), Mesorhizobium loti (26),Methanococcus jannaschii (27), Methanopyrus kandleri(28), Mycobacterium bovis (29), Mycobacterium tuberculosis(30), Mycoplasma genitalium (31), Myxococcusxanthus (32), Nanoarchaeum equitans (33), Prochlorococcusmarinus (34), Pseudomonas aeruginosa (35), Rickettsiaprowazekii (36), Sulfolobus solfataricus (37), Synechococcuselongatus (38), Thermoplasma acidophilum (39),Ureaplasma urealyticum (40) and Magnetococcus sp.(NC_008576) were used. We used needle, an implementationof the Needleman–Wu¨ nsch algorithm available in theEMBOSS package (41) for all sequence identity analyses.The algorithms discussed were implemented in PERL(script provided as Supplementary Data) on a Linuxplatform.Analysis for non-random codon usage dependencyon flanking amino acid residuesFor codons of interest, random occurrence model wasconstructed based on codon usage and amino acidfrequencies in a given genome. We used 10 000 suchrandom sets to calculate the z-scores for each residue–codon–residue combination. From the z-scores, P-valueswere calculated and were multiply corrected for bothcodon occurrence and amino acid occurrence biases usingBonferroni correction. To identify those combinations thathave a skewed occurrence, we used a stringent thresholdof P50.0001.Creation of the probability matrix for CSMCodon usage in the genome interest was calculated usingthe CUSP program in the EMBOSS package (41), and acodon usage probability table was created based on thatinformation. For each amino acid the segmented probabilityinterval spans from 0.0 to 1.0, where eachconsecutive non-overlapping segment corresponds toprobability of a unique codon (Figure 1A). This probabilityinterval matrix had 64 individual data points under21 categories (20 amino acids + stop codons).Creation of the probability matrix for NCMFor each tripeptide A1–A2–A3 in the genome of interest,we calculated the usage probabilities of codons for A2flanked by A1 and A3. Based of these probabilities, wecreated a probability interval matrix for all combinationsof A1–C –A3, where C is the codon that encodes A2. Theprobability interval matrix thus created had 24 400individual data points under 8000 categories(20 20 20 amino acid combinations). Creation ofsuch a probability interval for the tripeptide S–A–S isillustrated in Figure 1B.Reverse translationIn reverse translation using CSM, a random number r wasgenerated where 0 r 1, for each amino acid in thequery protein sequence. The codon corresponding to theprobability interval within which r falls was chosen forreverse translation. In NCM, overlapping tripeptides wereused instead of single codons, and the codon was predictedfor the second residue. However, when reverse translatingwith NCM, the first residue and stop codons were assignedbased on probability alone. This procedure is alsoillustrated in Figure 1. c-NCM was created as anenhancement to NCM, in which reverse translation wasperformed n times using NCM for each protein sequence.The final DNA sequence was obtained by creating aconsensus sequence from the n sequences created.
Page 3 and 4:
Statement of lengthThis thesis does
Page 5 and 6:
molecules also bind to enzymes dire
Page 7 and 8:
Table of ContentsTitlePagesIntroduc
Page 9 and 10:
TitlePageFigure 4.5: Catalytic and
Page 11 and 12:
IntroductionFunctional and comparat
Page 13 and 14:
egulation owing to the large amount
Page 15 and 16:
(A) Closed complex formationRNA pol
Page 17 and 18:
Intuitively negative supercoiling,
Page 19 and 20:
I.3. Transcription factors: domain
Page 21 and 22:
I.4. Transcriptional regulatory net
Page 23 and 24:
(Balazsi et al. 2005). Gene express
Page 25 and 26:
metabolism / carbon nutrition with
Page 27 and 28:
I.4.5. Signal sensing in transcript
Page 29 and 30:
I.5. Non-transcriptional output: cy
Page 31 and 32:
transcriptional initiation by c-di-
Page 33 and 34:
phosphatase activity of the kinases
Page 35 and 36:
I.7. ConclusionIn this chapter, we
Page 37 and 38:
Chapter 1: Global regulators of bac
Page 39 and 40:
Page 41 and 42:
Page 43 and 44:
Page 45 and 46:
Page 47 and 48:
Page 49 and 50:
Page 51 and 52:
Page 53 and 54:
Page 55 and 56:
Page 57 and 58:
Page 59 and 60:
Chapter 2: Transcriptional control
Page 61 and 62:
Page 63 and 64:
Page 65 and 66:
Page 67 and 68:
Page 69 and 70:
Page 71 and 72:
Page 73 and 74:
Page 75 and 76:
Page 77 and 78:
Page 79 and 80:
Page 81 and 82:
Page 83 and 84:
Page 85 and 86:
Page 87 and 88:
Chapter 3: Transcriptional and post
Page 89 and 90:
Page 91 and 92:
Page 93 and 94:
Page 95 and 96:
Page 97 and 98:
Page 99 and 100:
Page 101 and 102:
Page 103 and 104:
Page 105 and 106:
Chapter 4: Comparative genomics of
Page 107 and 108:
Page 109 and 110:
Page 111 and 112:
Page 113 and 114:
Page 115 and 116:
Page 117 and 118:
Page 119 and 120:
Page 121 and 122:
Page 123 and 124:
Page 125 and 126:
Page 127 and 128:
Page 129 and 130:
Page 131 and 132:
Page 133 and 134:
Page 135 and 136:
two-domain architectures wherethese
Page 137 and 138:
Page 139 and 140:
Page 141 and 142:
Page 143 and 144:
Page 145 and 146:
Page 147 and 148:
Page 149 and 150:
(A)60% of genomes from class4530150
Page 151 and 152:
Page 153 and 154:
Discussion and conclusionsDiscussio
Page 155 and 156:
Discussion and conclusionsD1.1.3. C
Page 157 and 158:
Discussion and conclusionsD2. Futur
Page 159 and 160:
BibliographyAmikam, D., Steinberger
Page 161 and 162:
BibliographyBlattner, F.R., Plunket
Page 163 and 164:
BibliographyChen, S.L., Hung, C.S.,
Page 165 and 166:
BibliographyEchols, N., Harrison, P
Page 167 and 168:
Bibliographysystem level analysis o
Page 169 and 170:
BibliographyHerring, C.D., Raghunat
Page 171 and 172:
BibliographyJenal, U., and Malone,
Page 173 and 174:
BibliographyKeseler, I.M., Collado-
Page 175 and 176:
BibliographyLee, S.H., Angelichio,
Page 177 and 178:
BibliographyMcClelland, M., Sanders
Page 179 and 180:
BibliographyO Croinin, T., Carroll,
Page 181 and 182:
BibliographyPrice, M.N., Dehal, P.S
Page 183 and 184:
BibliographySatriano, J., and Schlo
Page 185 and 186:
BibliographySteinberger, O., Lapido
Page 187 and 188:
BibliographyUssery, D.W., Wassenaar
Page 189 and 190:
BibliographyYanofsky, C. 1981. Atte
Page 191 and 192:
Appendix 1Appendix 1.1: Differences
Page 193 and 194:
!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!
Page 195 and 196:
Appendix 1Appendix 1.5: Gene expres
Page 197 and 198:
Appendix 2Appendix 2.1: Differences
Page 199 and 200:
Appendix 2Appendix 2.3: Co-regulati
Page 201 and 202:
Appendix 2Appendix 2.5: Flux coupli
Page 203 and 204:
Appendix 2Appendix 2.7: Co-regulati
Page 205 and 206:
Appendix 2Appendix 2.8B: Connectivi
Page 207 and 208:
Appendix 2Appendix 2.10: Correlatio
Page 209 and 210:
Appendix 2Appendix 2.12A: Robustnes
Page 212 and 213:
Appendix 3Appendix 3Supplementary m
Page 214 and 215:
Reaction ID Regulatory metabolite I
Page 216 and 217:
Appendix 3Appendix 3.3B: Direct reg
Page 218 and 219:
Appendix 3Appendix 3.4B: Direct reg
Page 220 and 221:
Appendix 4Appendix 4Supplementary m
Page 222 and 223:
Appendix 4Motile organisms are rich
Page 224:
Appendix 4We can expand the analysi
Page 227 and 228:
Appendix 5Appendix 5.1A: PFAMs for
Page 229 and 230:
Appendix 5Appendix 5.1C: PFAMs for
Page 231 and 232:
Appendix 5Appendix 5.2B: Small-mole
Page 233 and 234:
Appendix 5Appendix 5.3: Organisms s
Page 235 and 236:
Appendix 5Appendix 5.5: Partner dom
Page 237 and 238:
Appendix 5fraction of SDRRs would a
Page 239 and 240:
Appendix 5Appendix 5.6C: Non-hybrid
Page 241 and 242:
Appendix 5Appendix 5.8: List of org
Page 244 and 245: Appendix 6Appendix 6List of genomes
Page 246 and 247: Appendix 648. Mycobacterium tubercu
Page 248 and 249: Appendix 6144. Bacillus cereus ATCC
Page 250 and 251: Appendix 6240. Nitrobacter winograd
Page 252 and 253: Appendix 6336. Cytophaga hutchinson
Page 254 and 255: Appendix 6432. Acinetobacter bauman
Page 256 and 257: Appendix 6528. Enterobacter sakazak
Page 258 and 259: Appendix 6624. Polynucleobacter nec
Page 260 and 261: Appendix 7Appendix 7List of genomes
Page 262 and 263: Appendix 748. Bordetella parapertus
Page 264 and 265: Appendix 7144. Haemophilus somnus14
Page 266 and 267: Appendix 7240. Prochlorococcus mari
Page 268 and 269: Appendix 7336. Synechococcus sp. CC
Page 270 and 271: Downloaded from genome.cshlp.org on
Page 284 and 285: JOURNAL OF BACTERIOLOGY, June 2008,
Page 286 and 287: VOL. 190, 2008 P. MIRABILIS GENOME
Page 294: VOL. 190, 2008 P. MIRABILIS GENOME
Page 299 and 300: 4 Nucleic Acids Research, 2008Table
Page 301 and 302: 6 Nucleic Acids Research, 2008Table
Page 303 and 304: 8 Nucleic Acids Research, 2008solve
Page 305 and 306: 524 Mol Genet Genomics (2008) 279:5
Page 317 and 318: 4378 J.J. James et al. / FEBS Lette
Page 324 and 325: An Assessment of the Role of DNA Ad
Page 326 and 327: DAM and Transcriptionpotential of s
show all

A computational study of bacterial gene regulation and adaptation ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?