GN 415 GN 415
Popular in Course
Popular in Genetics (Graduate Group)
verified elite notetaker
This 7 page Class Notes was uploaded by Geraldine Kohler PhD on Thursday October 15, 2015. The Class Notes belongs to GN 415 at North Carolina State University taught by Ga Gibson in Fall. Since its upload, it has received 11 views. For similar materials see /class/223789/gn-415-north-carolina-state-university in Genetics (Graduate Group) at North Carolina State University.
Reviews for GN 415
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/15/15
ES ZT EPJ 39 quot F DF F 1 l39vquotL39sT 3 THE 3391quot h t lllil iSCEI39v39Efbi39GnfiaflCDlT Relative Impact of Nucleotide and Copy Number Variation on Gene Expression Phenotypes Barbara E Stranger et al Science 315 848 2007 DOI 101126science1136678 Science The following resources related to this article are available online at wwwsciencemagorg this information is current as of April 25 2007 Updated information and services including highresolution gures can be found in the online version of this article at httpwwwsciencemagorgcgicontentfull3155813848 Supporting Online Material can be found at httpwwwsciencemagorgcgicontentfull3155813848DC1 This article cites 27 articles 7 of which can be accessed for free httpwwwsciencemagorgcgicontentfull3155813848otherartices This article appears in the following subject collections Genetics httpwwwsciencemagorgcgicollectiongenetics Information about obtaining reprints of this article or about obtaining permission to reproduce this article in whole or in part can be found at httpwwwsciencemagorgaboutpermissionsdtl Science print ISSN 00368075 online ISSN 10959203 is published weekly except the last week in December by the American Association for the Advancement of Science 1200 New York Avenue NW Washington DC 20005 Copyright c 2007 by the American Association for the Advancement of Science all rights reserved The title SCIENCE is a registered trademark of AAAS Downloaded from wwwsciencemagorg on April 25 2007 REPORTS References and Notes K Rose M Whitt in Fields Virology D N Knipe P M Howley Eds Lippincott Williams amp Wilkins Philadelphia ed 4 2001 pp 1221 1244 2 B L Rao el ol toncel 364 869 2004 H a 3 i Le Blanc elol Nol Cell Biol 7 653 2 05 4 L H Luo V Li R M Snyder R R Wagner Virology 163341 1 88 A Benmansour el ol Virol 65 4198 1991 6 S B Vandepol L Lefrancois Holland Virology 148 312 1986 7 V Gaudin C Tuftereau D Segretain M Knossow A Flamand Virol 65 4853 1991 8 W Doms D S Keller A Helenius W E Balch Cell Biol 105 1957 1987 V Gaudin Subcell Biochem 34 379 2000 V Gaudin R W Ruigrok M Knossow A Flamand Virol 67 1365 1993 P Durrer V Gaudin R W Ruigrok R Graf Brunner Biol Chem 270 17575 1995 B L Fredericksen M A Whitt Virology 217 49 1996 C C Pak A Puri R Blumenthal Biochemislry 36 8890 1997 F A Carneiro A S Ferradosa A T Da Poian Biol Chem 276 62 2001 S Roche V Gaudin Virology 297 128 2002 A Flamand 5quot H ow b b b b F Fwa F He chm lt 9 g 3 m H E 8 a s H r S Roche S Bressanelli F A Rey V Gaudin Science 313 187 2006 M Kielian F A Rey Nol Rev Microbiol 4 67 2006 P A Bullough F M Hughson Skehel D C Wiley Nolure 371 37 1994 H W H o 20 Skehel D C Wiley Cell 95 871 1998 21 H S Wn R G Paterson X Wen R A Lamb T S ardetzky Proc Noll Acod Sci USA 102 9288 2005 V Niodis S Ogata D Clements S C Harrison Nolure 427 313 2004 S Bressanelli el ol EMBO 23 728 2004 D L Gibbons el ol Nolure 427 320 2004 E E Heldwein el ol Science 313 217 2006 V Gaudin R W Ruigrok C Tuftereau M Knossow A Flamand Virology 187 627 1992 Single letter abbreviations for the amino acid residues are as follows A Ala C Cys D Asp E Glu F Phe G le H H NNNN N anew N N r 0 G F Forster 0 Medalia N Zauberman W Baumeister D Fass Proc Noll Acod Sci USA 102 4729 2005 P Zhu el ol Nolure 441 847 2006 H S Vin X Wen R G Paterson R A Lamb l S ardetzky Nolure 439 38 2006 V Gaudin H Raux A Flamand R W Ruigrok Virol N a WM 9w w H R Glen K Diederichs 0 Kohlbacher i Fischer Eds Lecture Noles in Bioinformon cs vol 3695 Springer Verlag Berlin 2005 pp 163 1 4 33 F Lafay A Benmansour K Chebli A Flamand Gen Virol 77 339 1996 34 v Gaudin Virol 71 3742 1997 35 C Tutfereau Benejean D Blondel B Kieffer A Flaman EMBO 17 7250 1998 L V Chernomordik M M Kozlov ell 123 375 2005 37 C M Carr C Chaudhry P S Kim Proc Noll Acod Sci USA 94 14306 1997 38 W L Delano The PyNiOL Molecular Graphics System DeLano Scientific San Carlos CA 2002 available at vwwvp o or 39 We thank A Flamand for constant support on this project Lepault R Ruigrok M Knossow A Benmansour C Tunereau and D Blondel for helpful discussions at different stages of this worlc and C illiaheu for virus purification Data collections were performed at the Swiss Light Source SLS Paul Scherrer instituL Villigen Switzerland and at the European Syndirotron Radiation Facility ESRF Grenoble France We acknowledge the help of T Tomizaki beamline X06SA SLS G Leonard and D Bourgeois beamlines iD29 and iD23 2 ESRF and S Duquerroy and G Squires in data collection We adknowledge support from the CNRS and i A the CN RS program Physique et Chimie du Vivantquot the iNRA Animal health department program 39 des animaux et leurs interactions avec la cellulequot the Ministere de l education nationale de la recherdie et de blanche and the Agence Nationale de la Redierdie prog S R was the recipient of an Agence Nationale de Recherche sur le Sida fellowship during part of this project Coordinates and structure factors have been deposited with the Protein Data Bank under accession code 2j6j Supporting Online Material vwwvsciencemagorgcgicontentfull3 155813843DC1 References Movie S1 29 September 2006 accepted 3 January 2007 101126science1135710 Relative Impact of Nucleotide and Copy Number Variation on Gene Expression Phenotypes Barbara E Stranger1 Matthew S Forrest1 Mark Dunning2 Catherine E Ingle1 Claude Beazley1 Natalie Thorne2 Richard Redon1 Christine P Bird Anna de Grassi3 Charles LeeL5 Chris TylerSmith1 Nigel Carter1 Stephen W Scherer 7 Simon Tavar 2 8 Panagiotis Deloukas1 Matthew E Hurles1 Emmanouil T Dermitzakis 1 Extensive studies are currently being performed to associate disease susceptibility with one form of genetic variation namely singlenucleotide polymorphisms SNPs In recent years another type of common genetic variation has been characterized namely structural variation including copy number variants CNVs To determine the overall contribution of CNVs to complex phenotypes we have performed association analyses of expression levels of 14925 transcripts with SNPs an CNVs in individuals who are part of the International HapMap project SNPs and CNVs captured 836 and 177 of the total detected genetic variation in gene expression respectively but the signals from the two types of variation had little overlap Interrogation of the genome for both types of variants may be an effective way to elucidate the causes of complex phenotypes and disease in humans nderstanding the genetic basis of phe pic variation in human een interrogated in a variety of species and experi mental scenarios in order to investigate the genetic basis of variation in gene regulation 1 8 as well as to tease apart regulatory net works 9 10 In some respects a comprehen sive survey of gene expression phenotypes steady state levels of mRNA serves as a proxy for the breadth and nature of phenotypic var iation in human populations 1 Much of the observed variation in transcript levels may be compensated at higher stages of regu latory networks but an understanding of the nature of genetic variants that affect gene ex pression will provide an essential framework and mode for elucidating the causes of other types of phenotypic variation Single nucleotide polymorphisms SNPs have long been known to be associated with phenotypic variation either through direct causal effects or by serving as proxies for other causal variants with which they are highly correlated ie in linkage disequi librium 1 2 12 An understanding 0 this as sociation has been facilitatedby the validation of millions of SNPs y the International HapMap project 13 However during the last few years structural variants such as copy number variants CNVs defined as DNA segments that are 1 kb or larger in size present at variable copy number in com arison with a reference g ome 14 have attracted much attention 2 It has become apparent that they are quite common in the human genome 15 19 and can have dra matic phenotypic consequences as a result of altering gene dosage disrupting coding se Wellcome Trust Sanger institute Wellcome Trust Genome Campus Hinxton Cambridge CB10 15A UiC 2Department of Oncology University of Cambridge Cancer Research UK Cambridge Research institute Li Ka Shing Centre Robinson Way Cambridge CB2 ORE UK 3istituto di Tecnologie Biomediche Sezione di Bari Consiglio Nazio nale della Ricerche CNR 70126 Bari italy 4Department of Pathology Brigham and Women s Hospital and Harvard Medical School Boston MA 02115 USA SBroad institute assachusetts institute of Technology Cambridge MA 02142 USA 6The Centre for Applied Genomics and Program in Genetics and Genomic Biology The Hospital for Sick Children MaRS Centre Toronto Ontario MSG 1L7 Cana a 7Department o quences or perturbing long range gene regu lation 20 2 Evidence has been presented that increased cop number can be positively 18 22 or negatively 23 correlated with gene expres sion levels for example deletion of a transcrip tional repressor could serve to elevate gene expression but the relative contribution of such large genetic variants ie CNVs and smaller to phenotypic variation has not been evaluated It is also still unknown whether SNPs can serve as proxies to CNVs 24 25 and whether the complex nature of some CNVs requires that they be surveyed directly 26 We have used the phase I HapMap SNPs 13 and the recently described CNV data ascertained in the same HapMap populations 26 for correlation with genome wide gene expression variation in the same individuals ene expression was interrogated in lympho blastoid cell lines of all 210 unrelated HapMap individuals 13 from four populations CEU 60 Utah residents with ancestry from northern and Han Chinese in ba in Ibadan Nigeria in four technical rep licates see Methods Out of the 47294 tran scripts that were interrogated the normalized values for 14925 transcripts 14072 genes were included in the analysis see Methods and 2 7 The SNP genotypes from phase IHapMap 28 were used in the analysis see Methods CNV data were represented by log ratios from com parative genomic hybridization CGH of each HapMap individual against a common reference individual on an array comprising 26574 large insert clones covering 937 of the euchromatic portion of the genome 26 29 Log ratios from two sets of clones were analyzed the whole set of 24963 autosomal clones CGH clones and the 1322 autosomal clones corresponding to CNVs present in at least two HapMap individ uals CNV clones 26 We excluded genes on expected to be signi cant eg Bonferroni cor rection or by setting the threshold to a value that generates a satisfactory false discovery rate R We have used the second and we have estimated the FDR on the basis of the number of genes tested and have required that in all cases at least 80 of the genes called signi cant are estimated to be truly signi cant Given that there are 14072 genes that lie within 1 Mb of SNPs and within 2 Mb of the full set of CGH clones and 7150 genes that lie within 2 Mb from the CNV clones from 7135 to 7191 depending on the population owing to missing data we ex pect this analysis to generate false positive as sociation signals for approximately 14 and 7 genes respectively in eac population the 14072 genes tested we detected sig ni cant associations with at least one SNP for 323 348 370 and 411 genes for CEU CHB IPT and YRI respectively eg Table 1 and table S1 These comprise a total of 888 non redundant genes of which 331 37 were replicated at the same signi cance level in at least one other population and of those 67 8 were signi cant in all four populations Table 2 and table S2 As expected we have limited power to detect weak effects because of the small sample sizes The minimum detected squared regression coef cient 72 which re ects the proportion of expression variance accounted for by the linear association with allele counts was 027 However some very strong effects were detected that in some cases had an 72 close to 1 Fig 1 and g S1 We detected a strong preference for associated SNPs to be close to their respective genes most of which were within 100 kb of the interrogated REPORTS expression probe Fig 1 A and C In summary we detected a large number of regions that ap t carry genetic variation affecting gene expression To evaluate the effect of experirnen al varia 39on hence the robustness of our associations we compared the list of gene ex pression associations from our previous study 1 phenotypes 47 went into the current analysis o which 43 915 were called signi cant at the same permutation threshold 005 in the same population The previous study was performed with different batches of cells by using RNA ex 39 different laboratory with RNA levels quanti ed on a different type of array custom versus genomewide array so the high degree of experimental and statistical replication strongly suggests that the signals we detected are robust an stable to experimental variation in expression measurements Of the 14072 genes tested we detected sig ni cant associations with at least one of the 24962 autosomal CGH clones in 85 44 58 and 96 genes in CEU HB IPT and YRI re spectively 238 nonredundant genes of which 28 12 were replicated at the same signi cance level in at least one other population and of those 5 2 were signi cant in all our populations Fig 2 Table 1 and table S3 and S2 and S3 Not all associated clones were within CNVs de ned using the stringent criteria of26 119 out of303 39 associated clones were previously defined as CNVs it is like y that some of these clones encompass smaller CNVs that are detectable though asso Table 1 Numbers of genes with significant associations to SNPs SNPprobe distance lt 1 Mb all CGH clones cloneprobe distance lt 2 Mb or CNV clones cloneprobe distance lt 2 Mb as assessed by permutations together with the numbers of overlaps between SNPassociated genes and CGH or CNV cloneassociated genes probevariant distance lt 1 Mb for both SNPs and clones see table S4 sex chromosomes because of their imbalance in CNV 1 Mb males and females We performed linear regres Gene population CNV 2 Mb SNP SNP overlap 01 01 5301 Of e four POPUIa OI S separate CGH clones CNV clones CGH clones CNV clones etween quot J A 39 39 D r 39 I values and SNP genotypes or clone log ratios Permummn thTEShOId 001 that were near the gene SNP position or clone CEU 362 38 14 15 midpoint within 1 Mb and 2 Mb respectively of CH5 221 110 673 10 9 the probe midpoint position We used different JPT 319 134 752 13 14 window sizes for SNPs and clones because I 481 166 815 14 11 Clones are large median sin of N17 b and Nonredundant 1246 451 I 1886 28 16 structural variants can exalt long range e ems Permutation threshold 0001 2 so a 2 Mb window is more appropriate CEU 85 9 8 Statistical signi cance was evalu t th CHB 44 32 348 5 6 use of permutations 30 as prev1ously described JPT 58 4 37 8 6 1 and corrected v ue threshold of 0001 W 96 42 411 7 6 was applied see Methods Repeated permuta Nomadundam 238 9 I 888 15 12 tion exercises showed that our permutation Pe mmatm thTEShOId 00001 thresholds were very stable see table S1 We EU 32 18 198 5 6 tested a large number of genes so an additional CH5 14 19 204 4 4 correction was required This could be done JPT 23 20 217 6 5 either by adjusting the threshold to a new cor VRI 27 16 251 2 2 rected threshold above which all genes are 69 39 526 8 8 wwwisciencemagiorg SCIENCE VOL 315 9 FEBRUARY 2007 849 r o o 01 Lo 01 E Q lt C o O i O U m E m 0 C 9 0 V i g i E o c 390 m 390 m 2 C g o D REPORTS ciations of log ratios across a population but cannot be detected as extreme outliers in their log ratios in any one individual as is required for classi cation as a CNV in 26 see ex ample below For 36 common minor allele frequency gt 005 CNVs encompassing 99 CGH clones accurate CNV genotypes were available We used these genotypes to validate the statistical power of performing association analysis using log ratios directly ra er genotypes There wa strong correlation be tween 72 values or P values generated using the log ratio signals or the CNV genotypes Pearson correlation coef cients gt 09 indicat ing that log ratios can be used directly Little rior data exists on CNV expression associations against which to compare and demonstrate the robustness of our associations expression in a subset of the CEU Two of these deletions are covered by our CGH data The re ported expression association caused by the largest of these two deletions is also captured in our analysis in uencing UGT2B 7 and we extend this observation to show that this deletion also affects the expression of three other near genes UGHB 7 UGUBZ 0 and UGUBZZ and that these associations replicate across all four populations The smaller deletion of only 18 kb reported previously 18 as affecting expression of GST 39 the expected resolution of the CGH data Nonetheless we observe an as sociation that although it does not pass our stringent permutation threshold 0001 has signi cant nominal P values in all four pop ulations PCEU 00292 Pym 00018 Pm 00408 PCHB 00185 This suggests that effects of CNVs far smaller than genomic regions that met our criteria to be called a NV within the CGH platform can be detected and replicated in multiple populations with our analysis Having investigated the potential contribu tion of CNV to variation in gene expression by using data from all CGH clones we interrogated the nature of CNV effects on gene ex ression in ner detail by performing association tests of 1322 clones within high con dence CNVs see above with expression of the 14072 genes in order to generate a set of high stringency asso ciations for which the presence of an underlying CNV has already been validated Signi cant as sociations with at least one of the 1322 CNV clones were detected for 40 32 40 and 42 genes in CEU CHE IPT and YRI respectively 99 nonredundant genes table S4 Thirty four of the 99 genes 34 associated with CNV clones have a signi cant signal in at least two populations Table 2 of which 7 7 were as sociated in all populations Some CNV clones were associated with more than one gene in the same population a notable exarrrple was a single CNV clone associated with expression of four genes in all populations UGUB genes see above CNVs detected by CGH can be classi ed into ve classes deletion duplication deletion and duplication at the same locus multial elic and co lex 26 we find all classes of CNV represented among the signi cant associations Despite the clear preference for genes to lie close to their associated CNVs Fig 1 B andD 53 of the expression probes associated with a CGH clone were located outside the CNVs encom Table 2 Sharing of associations between populations CGH clone CNV clone SNP 2 Mb 2 Mb 1 Mb CEUCHBJPTVRI 5 7 67 CEUCHBJPT 4 48 CEUCHBVRI 1 0 11 EUPTVRI 1 0 12 CHBJPTVRI 3 3 28 CEUCHB 1 3 18 EUPT 2 0 15 CEUVRI 6 6 36 CHEPT 4 5 51 CHBVRI 1 3 18 PTVRI 2 3 27 EU only 67 20 116 CHB only 27 7 107 PT only 39 18 122 VRI only 77 20 212 Sum 238 99 888 Gene associations in at least two populations 28 34 331 Percentage of total 012 034 037 Gene associations in single populations 210 65 557 Percentage of total 088 066 063 passing that clone 26 This suggests that rather than altering gene dosage about half the CNV effects are caused by disruption of the gene some parts of the gene but not the probe are within in the CNV or affect regulatory regions and other functional regions that have an impact on gene expression When we extended our analysis to consider associations between genes and CNVs up to 6 Mb ap we detected a few signi cant long distance associations beyond 2 Mb table S5 These types of long range effects are becoming more apparent through recent studies looking in detail at speci c genomic regions 20 3 A small minority 5 to 15 of e signi cant CNV expression associations have anegative correlation between co ber and gene expression which suggests that not all the detected effects are of the conventional type wherein gene expression levels increase with gene copy number table S3 Almost all 32 e associations that are shared between populations also exhibit the same di rection of correlation in all populations The two exceptions could result from the CNVs being in linkage disequilibrium with different regulatory variants in different opulations or because of SNP gtlt CNV interactions However e strong bias toward positive correlations between copy number and expression levels implies that the vast majority of these associations are attrib utable to the CNV itself and not to a linked variant We next determined whether the same as sociations were also captured by SNPs Fig 2 and gs S3 and S4 We only considered those CGH clones or CNVs within 1 Mb of the probe so that the analysis is comparable to that of the SNPs total of 188 and 84 genes for CGH clones and CNVs respectively We expect some of the NVs to be correlated with SNPs via common genealogical history linkage dise uilibrium and therefore their effect on gene expression would also be captured by SNP associations Fewer than 20 in all populations of the de tected CGH clone associations overlapped with SNP associations Table 1 even when we in cluded CGH an SNP associations with the same gene but in different populations 28 out of 188 14 genes with signi cant CGH clone associations also had a SNP association in any population The same is true of CNV clone as sociations Only 15 of 84 genes 18 with CNV clone associations within 1 Mb also had a SNP association in any population and if we required the association in the same population only 12 14 of genes had a SNP association On the basis of previous work characterizing the patterns of linkage disequilibrium aroun s 26 we considered that this low overlap be tween CNVor CGH clone associations with SNP associations might be due in part either to a low density of successfully genotyped SNPs around duplications SDs are the primary cause of low 9FEBRUARY2007 VOL 315 SCIENCE www5ciencemagorg Downloaded from wwwsciencemagorg on April 25 2007 SNP densities in HapMap Phase Ibecause of the dif culties in developing robust SNP genotyping assays within them 13 We did not observe en richrnent of segrnentally duplicated sequences within the CGH and CNV clones at n share signals with SNPs relative to those CGH and CNV clones that did share signals with SNPs However we observe a 25 fold excess of compound CNVs CNVs with more than one mutation event on the basis of classi cation of the CNVs in 26 in associations that are not shared with SNPs relative to those that are shared Fisher s exact test P lt 0001 Thus our analysis suggests that recurrent mutation is a likely factor reducing overlap between CNV and NP a socia 39ons CNV associations that were also detected with SNPs were clearly biased toward large effect sizes tables Sl and S3 Of the 12 genes with both SNP and CNV associations in the same population 8 shared the association in two or more populations giving a redundant total across the four populations of 26 shared CNV NP associations The ratio of 8 out of 12 67 population shared associations is larger 0 250000 than that observed in all CNV associations 34 out of 99 34 potentially suggesting that associations with higher frequency older are more likely to be captured by SNPs For the 26 associations representing 12 genes see above captured both by CNVs and SNPs in the same population we observed that SNPs and CNVs were themselves highly correlated for 23 out of 26 SNP CNV pairs Pearson correlation P lt 0 01 suggesting that for these cases the CNVand SNP captured the same effect and that only a small fraction of the associations captured both by SNPs and CNVs occurs by chance In summary 87 out of 99 87 of genes with a significant CNV association are not associated with SNPs m The large scale typically gt 100 kb copy number variation analyzed here appears to be associated with about 10 to 25 as many gene expression phenotypes as captured by 700000 SNPs an the majority of these effects cannot be explained by altered dosage of the entire gene but by gene disruption and its impact on the regulatory landscape of the region where these CNVs occur When we restrict the analysis to log10pvalue 0 u 0 1 o 0 250000 500000 7500001000000 Distant e Frequency 01 040506 10 Adjusted2 Fig 1 Strength of association as a function of distance between A SNP and probe and B CNV and probe Positive associations between mRNA levels and clone logz ratios are shown in red negative associations in black Distance equal to zero corresponds to the probe residing within the 500000 750000 1000000 40 r 30 e 20 r 10 It I 2 g 0 s a a 2 Frequency 01 REPORTS within 1 Mb of the probe of the expressed gene we detected 1061 genes associated with CGH clones or SNPs 177 of which are associated with CGH clones 836 with SNPs and 13 with both Of the 972 genes associated with CNV clones or SNPs 875 are associated with CNV clones 925 with SNPs and 125 with both Whereas the phase I HapMap SNPs likely capture a large fraction of the SNP effects in the genome 13 only a small minority of the CNVs in the genome were considered here CNVs lt 100 kb in length are far more numerous than CNVs gt100 kb in length 19 As a con sequence 875 to 177 is a minimal estimate of the proportion of heritable gene expression variation that is explained by copy number variation Our study has attempted to evaluate the rel ative impact of CNVs and SNPs on phenotypic variation in human populations Within the limi tations of our samples tissue type SNP cover age and CNV resolution each type of genetic variation captures a substantial number of large ly mutually exclusive effects on gene expres sion We also demonstrate that both CNV and t 0 Adjusted 2 CNV In each population panel only the details for the most significant association per significant gene are shown Distribution of r2 values for the most si nificant association per significant gene for C SNP expression associations and D cloneexpression associations www5ciencemagrorg SCIENCE VOL 315 9FEBRUARY2007 Downloaded from wwwsciencemagorg on April 25 2007 REPORTS 852 A 10 0 8 A7 2 00 I Lug 034 a a 2 1 10 0 8 A7 2 0 00 I S 5 034 a a 2 1 10 0 8 A7 2 0 i 035 2 m4 9 3 2 1 10 0 8 37 age gt 5 34 3 2 9 lt0 lt0 2 a x LU 0 08500000 00000000 00500000 70000000 70500000 71000000 71500000 Mme 9 lt0 lt0 2 a x LU o 08500000 00000000 00500000 70000000 70500000 71000000 71500000 coordinate Expression 0 r E 08500000 00000000 00500000 70000000 70500000 71000000 71500000 coordinate Expression 07 08500000 00000000 00500000 70000000 70500000 71000000 71500000 coordinate Fig 2 Examples of SNPsexpressiun and rioheexpressroh assudaiiuns in 1he roor HapMap popoieiiohs chromosome 5 p ressiun are expression and rioheexpre ooheexpressioh ass dir 5 Sighihrehi essorieiions betw observed in CEU CH3 and PT 001 hoi in W B SNPV ssioh ossorieiioh ror GBP3 hr 1 Bo uda un for SMNZ een dun d clones are signi can y assuciaied wiih expression of GBP3 in CEU CH3 and PT 001 hoi in W in ea ch plm ooheo lines show 1he 0001 permoieiioh 9 FEBRUARV 2007 VOL 315 Aloglqpovalue 4091qu lue logl 0p vaue 2 g E m s of ihe hum SCIEN CE easi WU Expression 0 z lt i 0 0000 Expression Expression HapMap individuals an genome insei panels show no SNP gehorypes or Clone lugz reiiosror1he mosi sighimerri Clone or SNP in mm popoieiioh which may oirrer across popoieiiohs wwwsc39iericemagorg 00000000 2007 Downloaded 39om www50iencemagorg on April 25 SNP associations are replicated across popula tions Replication of association signals is the sine qua non of association studies and the fact 39 div pop ulations and with small sample sizes highlights the relevance and robustness of the associations we detect Gene expression is the basis for many crucial functions in the cell so the relative con tribution of these two types of variants is an in dication of the nature of the mutational and natural selection processes that contribute to phenotypic diversity and divergence It is there fore essential thatwe interrogate both SNPs and CNVs f all types to perform a comprehensive exploration of genetic effects on phenotypic var iat39o and disease It is possible that ifa larger number of SNPs were analyzed or a higher reso lution of CNVs was available we would o serve more overlap between the effects attri models of association in small samples so it is very likely that if we apply more complex and realistic models e g epistatic interactions andor larger population sarnp es a ar er num ber of effects would be revealed The results presented here reinforce the idea that the com plexity of functionally relevant genetic variation ranges from single nucleotides to me gabases and the full range of the effects of all of these vari ts will be best captured and interpreted by complete knowledge of the sequence of many human genomes Until this is possible we need to survey all known types of genetic variation to maximize our understanding of human evolution diversity and disease A 0 References and Notes B E Stranger et at PtoS Genet 1 e78 2005 V G Cheung et at Nature 437 1365 2005 3 S Doss E E Schadt T A Drake A Lusis Genome Res NH buted to CNVs and SNPs However the di iculty 15 651 2005 of designing robust SNP genotyping assays in R B Brem L KruglyakProc Notl Acod Sci USA 102 structurally dynamic regions of the genome 26 1572 2005 q suggesm that even with more comprehensive 5 gdsitorey Ni Akey L Kruglyak PtoS Biol 3 e267 inteIrOga OH 0f SNPS and CNVS the OY laP 6 M F Oleksiak 1 L Roach a L CrawfordNot Genet may not be high enough for one type ofvariation 37 67 2005 to be suf cient for exploring the genetic causes 7 5 A Monks etal Am Hum Genet 75 1094 2004 of disease We have also demonstrated that it is 5 E E 5mm amquot Name 422 297 Qom 9 E Chesler et at Not Genet 37 233 2005 not necessary to PETfOIm 11013 Strum fmh CNV 10 L Bystrykh et at Not Genet 37 225 2005 calls or CNV genotypes but it is possible to use 11 E r Dermitzakis B E Stranger Momm Genome 17 filtered CGH log ratios or any other type of 503 21006 gt hi quality quantitative signal that re ects un 1239 Bergman 139 139 HUdm39 506quot 30639 647 de ymg CNV has also becomiaPParem that 13 international HapNiap Consortium Nature 437 1299 there are many more structural variants that con 2005 tribute tophenotypic variation than our stringent 14 L Feuk C R Marshall R F Wintle S W Scherer Hum criteria forwhat is a CNVreveal and thathigher M Genef 15 SUPPL Dr R57 200 15 A iatrate etolNot Genet 36 949 2004 resolution methods are necessary to elucidate 16 Sebatemp Sdeme 305 52 A 6 6 and mctlonLast butnotlefist IS 17 E Tuzun etol Not Genet 37 727 2005 the fact that we have only cons1dered Simple 18 S A NicCarroll etol Not Genet 38 86 2006 REPORTS 19 D F Conrad T D Andrews N P Carter Ni E Hurles K Pritchard Not Genet 38 75 2006 20 P Stankiewicz in Genomic Disorders The Genomic Bosis ofDiseose R Lupski P Stankiewicz Eds Humana Press Totowa NJ 2006 PP 357 3 9 D A Keinjan V van HeyningenAm Hum Genet 76 8 2005 Ni Somerville et at N Engl Med 353 1694 2005i A Lee etolAnn Neurol 59 398 2006 D P Locke etolAm Hum Genet 79 275 2006 D A Hinds A P Kloek Ni en X Chen K A Frazer Not Genet 38 82 2006 R Redon et at Nature 444 444 2006 GENEVAR GENe Expression VARiation vwwvsangerac N N ukgeneyar international Hapiiliap Project wwwhapmaporg release 16c 1 The Copy Number Variation CNV Project Data index vwwvsangeracukhumgench ata 30 R Doerge G A Churchill Genetics 142 285 1996 G Nierla etolAm Hum Genet 79332 2006 We thank A Clark and Pritchard for comments on earlier versions of the manuscript M Smith for assistance with software development and Ni Gibbs Orwick and C Gerringertor technical support Funding was provided by the Wellcome Trust to E T D All E H PD CTS and N ww NH aOntar io olarotthe Canadian institutes of Health Research and the Howard Hughes Medical institute Supporting Online Material vwwvsciencemagorgcgicontenttull3155813848DC1 Materials and Methods Figs 51 to 53 Tables 51 to 55 References and Notes 24 October 2006 accepted 5 January 2007 101126science1136678 Evidence That Focal Adhesion Complexes Power Bacterial Gliding Motility Tam Mignot1oshua W Shaevitz2 Patricia L Hartzell3 David R Zusmanlquot The bacterium Myxococcus xanthus has two motility systems S motility which is powered by type IV pilus retraction and A motility which is powered by unknown mechanisms We found that A motility involved transient adhesion complexes that remained at fixed positions relative to the substratum as cells moved forward Complexes assembled at leading cell poles and dispersed at the rear of the cells When cells reversed direction the Amotility clusters relocalized to the new leading poles together with Smotility proteins The Frz chemosensory system coordinated the two motility systems The dynamics of protein cluster localization suggest that intracellular motors and force transmission by dynamic focal adhesions can power bacterial motility two separate but coordinated motility engines S motility is powered by type IV pili that are as sembled at the leading cell pole movement is produced as the pili bind to surface exopolysac charides and are retracted thereby pulling the cell forward 2 A motility on the other hand is not uring the exhibition of gliding motility bacteria move across solid surfaces with out the use of agella 1 Gliding motility is important for biofilm formation and erial virulence Motility in Myxococcw xtmthus a Gram negative rod shaped bacterium relies on associated with pili or other obvious structures and is not well understoo To investigate the A motility system we studied Ang a protein that is essential for A motility but dispensable for S motility fig S1 A and B 3 Ang is similar to FrzS an S motility protein that oscillates from one cell pole to the other when cells reverse direction 4 g S1A To track the localization of Ang in moving cells we constructed an M xtmthus strain containing a chimeric 1ng 9 gene in place of the endoge nous 1ng gene g S2A This chimeric gene encodes an Ang yellow uorescent protein YFP fusion protein that was stable and func tional g S2 B and C We followed Ang YFP localization using time lapse video microscopy In fully motile cells Ang YFP was localized in 1Department of Molecular and Cell Biology University of California Berkeley CA 94720 2Department of integrative Biology University of California Berkeley CA 94720 USA Department of Microbiology Molecular Biology and Biochemistry University of idano Moscow D 83844 USA correspondence should be addressed E mail tmignotberkeleyedu TJVL zusman be rkeleyedu DRZi www5ciencemagtorg SCIENCE VOL315 9FEBRUARY2007 853 Downloaded from wwwsciencemagorg on April 25 2007
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'