AutoSeqMan: batch assembly of contigs for Sanger sequences

2018·1.2 MB·

by  Jin, Jie-Qiong

ZOOLOGICAL RESEARCH
AutoSeqMan: batch assembly of contigs for Sanger
sequences
Jie-QiongJin1,Yan-BoSun1,*
1StateKeyLaboratoryofGeneticResourcesandEvolution,KunmingInstituteofZoology,ChineseAcademyofSciences,KunmingYunnan
650223,China
ABSTRACT it is not suitable for many population studies due to high
costs and other limiting factors. For example, errors are
With the wide application of DNA sequencing
always introduced in final assembly and/or annotation results
technology, DNA sequences are still increasingly
using NGS data (Bickhart et al., 2017), and thus variations
generated through the Sanger sequencing platform.
detected in high-throughput analyses require validation by
SeqMan(intheLaserGenepackage)isanexcellent Sanger sequencing (Wall et al., 2014). Furthermore, for
programwithaneasy-to-usegraphicaluserinterface some present population genomic studies, error rates have
(GUI)employedtoassembleSangersequencesinto been found to increase with increasing depth of coverage for
Illuminadata,andthuscautionisneededwheninterpretingthe
contigs. However, with increasing data size, larger
resultsofnext-generationsequencing-basedassociationstudies
sample sets and more sequenced loci make contig
(Walletal.,2014). As such, Sanger sequencing technology is
assemblecomplicatedduetotheconsiderablenumber
stillwidelyusedinmanyresearchfields,includingevolutionary
ofmanualoperationsrequiredtorunSeqMan. Here, taxonomybasedonshortDNAsequences(Chenetal.,2017),
wepresentthe‘autoSeqMan’softwareprogram,which evolutionary history study of wild animals (Yuan et al., 2016),
can automatedly assemble contigs using SeqMan biodiversityestimatesandinfluencingfactors(Zhouetal.,2017),
and validation of mutations identified from high-throughput
scripting language. There are two main modules
analyses(Sunetal.,2013).
available, namely, ‘Classification’ and ‘Assembly’.
Further, with the wide application of DNA sequencing
Classification first undertakes preprocessing work,
technology, e.g. DNA barcoding, which uses short and
whereas Assembly generates a SeqMan script to standardized DNA sequences for individual identification of
consecutively assemble contigs for the classified organisms (Hajibabaei et al., 2007; Savolainen et al., 2005),
files. Throughcomparisonwithmanualoperation,we Sanger sequencing data are continuing to be accumulated
among evolutionary taxonomists and others. Thus, batch
showedthatautoSeqMansavedsubstantialtimeinthe
manipulation of these Sanger sequences has become an
preprocessing and assembly of Sanger sequences.
important task before downstream analyses, especially for
We hope this tool will be useful for those with large those who doesn’t have programming or bioinformatics
sample sets to analyze, but with little programming background or experiments. Although several sequence
experience. Itisfreelyavailableathttps://github.com/ manipulationpackagesforgeneralpurposeissueshavebeen
published previously, including MEGA (Kumar et al., 2016),
Sun-Yanbo/autoSeqMan.
EMBOSS(Riceetal.,2000),andFasParser(Sun,2017),these
packages are all based on assembled contigs (a consensus
Keywords: Batch processing; Sanger sequences;
Contigassembly;SeqMan
INTRODUCTION Received:21December2017;Accepted:08February2018
Foundationitems:Thedevelopmentofthispackagewassupportedby
DNA sequencing technology has experienced a revolutionary
theNationalNaturalScienceFoundationofChina(31671326)andthe
shift from automated Sanger sequencing (Sanger et al., 1977)
to next-generation sequencing (NGS; reviewed by Shendure YouthInnovationPromotionAssociation,ChineseAcademyofSciences
& Ji (2008) and Shendure et al. (2004)). Although NGS *Correspondingauthor,E-mail:[email protected]
has dominated due to its high throughput (Schuster, 2008), DOI:10.24272/j.issn.2095-8137.2018.027
SciencePress ZoologicalResearch39(2):123-126,2018 123
Biology, 65(5): 824–842.
Zhou WW, Jin JQ, Wu J, Chen HM, Yang JX, Murphy RW, Che J. 2017. Mountains
too high and valleys too deep drive population structuring and demographics in a
Qinghai-Tibetan Plateau frog Nanorana pleskei (Dicroglossidae). Ecology and
Evolution, 7(1): 240–252.

FIGURE LEGENDS
Figure1OverviewofassemblingtasksforSangersequences
CommontasksforassemblingSFaingguerres e1q Ouevnecrevsiienwclu odfe a(1ss)ecmlasbsliifnyign gtassekqus efnocre Sfialensginerto sceoqrrueesnpocensd ingfolders(alwayscompletedmanually)and(2)opening
SeqMantoaddtheab1filesthatCboemlomngotno atassakms efosra maspsleemanbdlitnhge nSaasnsgeemr bsleinqgutehnecmes( nienecdluindgec 1on) scidlaesrsaibflyeimngo usseequopeenrcaeti ofnilse)s. TheautoSeqManprogramwas
designedbasedonthesesteps,inwtioth cthoerr‘eCslapsosnifidciantgio nfo’aldnedr‘sA s(aselwmabylys’ fcuonmctipolnestecdo rrmesapnounadlilnyg) taonsdte 2p)2 oapnedn3in,rge sSpeeqcMtivealny. to
add the ab1 files that belong to a same sample and then assembling them (needing
regionofoverlappingDNAcsoengsimdeernatbsl)e amndounsoe koepyercaotinosnisd)e. rTahtieo nautoSeqCMLaAnS pSrIoFgIrCamAT wIOaNs designed based on
hasbeentakenonthebattchhesaes ssteepmsb, lwyiothf tShae n‘gCelarsssiefqicuaetinocne’ san.d ‘Assembly’ functions corresponding to step 2
This function is designed to automatedly create sub-folders
SeqMan is a popular program in the LaserGene software
and 3, respectively. according to the sample ID and/or sequenced locus. All
package (DNAStar, Inc., Madison, WI, USA), which is used
downstream analyses are performed in the corresponding
for assembling Sanger sequences into contigs and has been
sub-folders, where all analyzed results are also saved.
widelyappliedinagreatnumberofstudies. Itcanhandletwo
According to our laboratory experience, this is an efficient
to thousands of Sanger sequences at one time but requires a
andconvenientwaytomanageandquerylaboratorysamples
considerablenumberofmanualoperations(e.g.,mouseactions,
(Chenetal.,2017;Zhouetal.,2017).
Figure 1) to run. Hence, it is complicated and time-expensive
The only input is the directory name, which contains the
for those with large sets of samples to assemble. Fortunately,
raw ab1 files. There are several input prerequisites required
sincethereleaseofVersion7,SeqMannowprovidesascripting
for Classification performance. First, all files must be stored
language,includingcommandsforopening,naming,saving,and
in a same directory. Second, all files must be named
closing projects, and a single script may be used to execute
according to a certain pattern, i.e., “sample-locus-others”.
multipleassembliesconsecutivelywithoutmanualintervention.
For example, the file name “YPX24212_16S-2215_TSS
Here, we developed a program called autoSeqMan, which
20171122-0871-1171_H02.ab1” denotes that it is a DNA
provides a simple way to automatedly classify Sanger
sequenceof16Sandthesamplenumberis“YPX24212”. The
sequences and then consecutively assemble them on a
program will automatedly recognize the filename according
personalcomputer. Itismainlydesignedforresearcherswith
to the user-specified delimiter and then create a sub-folder
largesetsofsampleswithoneormorelocisequenced.
“YPX24212_16S”inthemainoutputfolder. Thedelimitercan
IMPLEMENTATIONANDREQUIREMENTS be “-“, “_”, or other. After classification, the program will list
allsub-folders,andtheusercanlookatthefilesclassifiedinto
autoSeqMan was developed into a standalone Windows
eachsub-folderbysimplyclickingthefoldername(Figure2).
desktop application (compiled and tested in Windows 7/10).
It involves two modules, ‘Classification’ and ‘Assembly’, ASSEMBLY
correspondingtosteps2and3inFigure1,respectively. Each
This function will automatedly assemble the classified
modulecanhandlemultiplefilesandneedstheusertoselect
sequencefiles. Itwillfirstreadthelistofclassifiedsub-folders
the directory either containing the raw Sanger sequence files
created by the ‘Classification’ function, and then generate a
(*.ab1files)orcontainingtheclassifiedsub-folderscreatedby
SeqMan script for consecutively assembling the sequences
‘Classification’. Theoretically,thereisnolimittothenumberof
in each sub-folder. To perform this function, the user must
filesthatcanbeanalyzed.
first install the LaserGene package (version 7 or higher), and
This tool requires the sequence files be named in a
then tell autoSeqMan the full path of the SeqMan program
specialized format, in which the sample IDshould be present
(whichcanbealwaysrecognizedautomatedlybyautoSeqMan),
atthebeginningofthefilename.TheClassificationmodulewill
after which the program will complete all assembly tasks and
recognizethesampleIDbytheappropriatedelimiterandthen
savetheassemblyresultsautomatedly. Thedefaultscriptwill
createsub-folders(seebelow). Forconvenience,autoSeqMan
generateallSQD,FAS,andSEQresults(Figure3).
also provides a “Rename” tool to help users rename the ab1
filesforthebelowanalyses.
124 www.zoores.ac.cn
Figure 2 Overview of ‘Classification’ function in autoSeqMan

To perform this function, user needs to tell autoSeqMan the full path of a directory
Figure2OverFviiegwuorfe‘ C2l aOssvifiecravtiioenw’ founf c‘tCiolnaisnsaifuitcoaSteiqoMna’n function in autoSeqMan
which contains all raw ab1 files. The sample name should be present at the beginning
Toperformthisfunction,userneedstotellautoSeqManthefullpathofadirectorywhichcontainsallrawab1files.Thesamplenameshouldbepresentatthe
beginningoftheTfiloe npameref,owhrimch wtihollfib set hrfeeu cofniglcen tizineoadmna,en ,d uwesxhetrirac hctn ewdeiebllyd tsbhe e taorue tcotoSegelnlqi zMeaadun taaoncdSco eredqxintMrgactaotentdh etbshyp eet chiffieue dlalud teoplSimaetqitheMr .aonf a directory
according to the specified delimiter.
which contains all raw ab1 files. The sample name should be present at the beginning
of the file name, which will be recognized and extracted by the autoSeqMan
according to the specified delimiter.

Figure3DefaultSeqManscripFtigfuorre a3s Dseefmaubllti nSgeqSMaanng secrrispetq foure anscseemsbling Sanger sequences
ThisscriptwillbegeneratedautomaTtehdisb sycariuptto wSeilql Mbea ng.eTnheerareteids anuotonmeeadtefdo rbuys eaurstotoSeeqdMitiat.n.I nTdheefraeu ilts, naoll SneQeDd, fSoEr Qus,earnsd toF ASoutputsaregenerated.
edit it. In default, all SQD, SEQ, and FAS outputs are generated.
PERFORMANCE the running efficiency of SeqMan. In this test, the Assembly

module required 64 s to consecutively assembly contigs for
The main aim of autoSeqMan is to save manual operation the classified sequences, also substantially less than the ~2
in preparing files and running the SeqMan program. To hrequiredformanualoperation,suggestingthesignificanceof
evaluateitsperformance,weappliedthistooltoourlaboratory autoSeqManindealingwithlargedatesets.
data (Chen et al., 2017; Zhou et al., 2017). In this test,
one hundred samples were used, each of which had two LIMITATIONS

ab1 files available. Results showed that the Classification
operation creFatiegdursueb 3-f oDldeefrsau(nlta mSeedqMsamanpl escIrDipats fwoerl laasssembIltinisg imSpaonrgtaenrt stoeqnuoteentchaets autoSeqMan does not undertake
locus name if provided) and moved the appropriate files into anyfiltrationmanipulationonthesequencedata,eventhough
thesub-folderTshwiisth sincr8ips,t swubisllt abnetia gllyenleesrsattheadn athuetotimmeatuesded by auptoooSre-qquMaliatyn.s eTqhueernec eise nndos naereeda lfwoary susperersse tnot. Thus, after
formanualoperation(about1h,astestedbyourcolleagues). running autoSeqMan, users should undertake quality control
edit it. In default, all SQD, SEQ, and FAS outputs are generated.
Performance of the Assembly operation greatly depended on measuresofthefinalassemblywithSeqMan. Inaddition,the

ZoologicalResearch39(2):123-126,2018 125
outputFastafileswillhaveverylongIDs,whichmightintroduce Kumar S, Stecher G, Tamura K. 2016. MEGA7: Molecular evolutionary
some errors in subsequent sequence analyses. If necessary, genetics analysis version 7.0 for bigger datasets. Molecular Biology and
userscanusethe“Sort&Rename”functionofFasParser(Sun,
Evolution,33(7):1870–1874.
2017)toshortentheseIDs.
Rice P, Longden I, Bleasby A. 2000. EMBOSS: the european molecular
biologyopensoftwaresuite.TrendsinGenetics,16(6):276–277.
COMPETINGINTERESTS
Sanger F, Nicklen S, Coulson AR. 1977. DNA sequencing with
Theauthorsdeclarethattheyhavenocompetinginterests. chain-terminating inhibitors. Proceedings of the National Academy of
SciencesoftheUnitedStatesofAmerica,74(12):5463–5467.
AUTHORS’CONTRIBUTIONS
SavolainenV,CowanRS,VoglerAP,RoderickGK,LaneR.2005.Towards
Y.B.S.andJ.Q.J.designedthestudyandwrotethemanuscript.Y.B.S.wrote writing the encyclopedia of life: an introduction to DNA barcoding.
thesoftware. J.Q.J.evaluatedtheperformanceofautoSeqMan. Allauthors Philosophical Transactions of the Royal Society B: Biological Sciences,
readandapprovedthefinalmanuscript. 360(1462):1805–1811.
SchusterSC.2008.Next-generationsequencingtransformstoday’sbiology.
ACKNOWLEDGEMENTS
NatureMethods,5(1):16–18.
Specialthankstoourcolleagues,Hong-ManChen,Yu-QiHe,andFangYan, Shendure J, Ji H. 2008. Next-generation DNA sequencing. Nature
fortheirhelpfulsuggestionsontheimprovementofautoSeqMan. Biotechnology,26(10):1135–1145.
ShendureJ,MitraRD,VarmaC,ChurchGM.2004.Advancedsequencing
REFERENCES
technologies:methodsandgoals.NatureReviewsGenetics,5(5):335–344.
Bickhart DM, Rosen BD, Koren S, Sayre BL, Hastie AR, Chan S, Lee J, Sun YB. 2017. FasParser: a package for manipulating sequence data.
LamET,LiachkoI,SullivanST,BurtonJN,HusonHJ,NystromJC,Kelley ZoologicalResearch,38(2):110–112.
CM, Hutchison JL, Zhou Y, Sun JJ, Crisà A, De León FAP, Schwartz Sun YB, Zhou WP, Liu HQ, Irwin DM, Shen YY, Zhang YP. 2013.
JC, Hammond JA, Waldbieser GC, Schroeder SG, Liu GE, Dunham MJ, Genome-widescansforcandidategenesinvolvedintheaquaticadaptation
ShendureJ,SonstegardTS,PhillippyAM,VanTassellCP,SmithTP.2017. ofdolphins.GenomeBiologyandEvolution,5(1):130–139.
Single-moleculesequencingandchromatinconformationcaptureenablede Wall JD, Tang LF, Zerbe B, Kvale MN, Kwok PY, Schaefer C, Risch N.
novo referenceassemblyofthedomesticgoatgenome.NatureGenetics, 2014.Estimatinggenotypeerrorratesfromhigh-coveragenext-generation
49(4):643–650. sequencedata.GenomeResearch,24(11):1734–1739.
ChenJM,ZhouWW,PoyarkovNA,Jr., StuartBL,BrownRM,LathropA, YuanZY,ZhouWW,ChenX,PoyarkovNA,Jr.,ChenHM,Jang-LiawNH,
WangYY,YuanZY,JiangK,HouM,ChenHM,SuwannapoomC,Nguyen ChouWH,MatzkeNJ,IizukaK,MinMS,KuzminSL,ZhangYP,Cannatella
SN,vanDuongT,PapenfussTJ,MurphyRW,ZhangYP,CheJ.2017.A DC,HillisDM,CheJ.2016.Spatiotemporaldiversificationofthetruefrogs
novelmultilocusphylogeneticestimationrevealsunrecognizeddiversityin (GenusRana): Ahistoricalframeworkforawidelystudiedgroupofmodel
Asianhornedtoads,genusMegophryssensulato(Anura: Megophryidae). organisms.SystematicBiology,65(5):824–842.
MolecularPhylogenetics&Evolution,106:28–43. Zhou WW, Jin JQ, Wu J, Chen HM, Yang JX, Murphy RW, Che J. 2017.
HajibabaeiM,SingerGA,HebertPD,HickeyDA.2007.DNAbarcoding:how Mountains too high and valleys too deep drive population structuring
itcomplementstaxonomy,molecularphylogeneticsandpopulationgenetics. and demographics in a Qinghai-Tibetan Plateau frog Nanorana pleskei
TrendsinGenetics,23(4):167–172. (Dicroglossidae).EcologyandEvolution,7(1):240–252.
126 www.zoores.ac.cn