轉(zhuǎn)錄組star比對(duì)分析流程講解22瑞客

2(d)不基于比對(duì)，

&counting。(2x150

(.fastq

(.fafile)

(.gtffile)(可選)

(2x150

Read

(.fastq

(.fafile)

(.gtffile)

(2x150

(.fa (.gtf (.fastq

FASTQqualityscores=ASCIIqualityscore-either33orIlluminaFASTQqualityscoresQ=-10log10(e)e=estimatedprobabilityofthebasecallbeingQ40:1errorin10,000basecalls;Q30:1errorinQ28:1.6errorin1000;Q25:3errorin1000;Q20:1errorin (2x150(.fastq 使用NCBI提供的SRA-toolkit中的工具fastq-dump直接SRR文件，并轉(zhuǎn)換雙端就自動(dòng)拆分，如果是單端不受的數(shù)據(jù)集一般比較大，放入 (nohupcmd&)。fastq- - --split---&fastq- - --split---& 有e。 fastqc x-axis:Positionin y-axis:Quality (箱線圖通過最大值、上四分位示質(zhì)量得分。將所有reads的第一位堿基質(zhì)量得分進(jìn)行箱線 每個(gè)readGC含量的分布圖。橫坐標(biāo)表示平均GC含量，縱坐標(biāo)表示reads數(shù)。左圖GC含量雙峰，表示樣品可能存在特定的序列污染如混入了引物二聚體，當(dāng)這一 ../FastQC/Configuration/adapter_list.txtFastQC工具會(huì)與這兩個(gè)文件里的序列進(jìn)行比 multiqc–d.–o 直方圖展示有無量異常高或低的樣本。橫軸表示reads數(shù)，縱軸表示給定reads數(shù)區(qū)間內(nèi)樣本數(shù)目，即給定深度有多少樣本。reads數(shù)低于20M

橫軸是GC含量的判斷，縱軸是堿基

Commondjava-jartrimmomatic-0.30.jarPE--phred33input_forward.fqinput_reverse.fqoutput_forward_paired.fqoutput_forward_unpaired.fqoutput_reverse_paired.fqoutput_reverse_unpaired.fq TRAILING:20MINLEN:36Twoinputfilesandfouroutputorsw o，

ismatchesinthe'se1d' GTF GTFGTF(GeneTransferFormat,GTF2.2)isanextensionto,andbackwardcompatiblewith,GFF2(GeneralFeatureFormat). GTF1-seqname-nameofthechromosomeorscaffold;chromosomenamescanbegivenwithorwithoutthe'chr'prefix.Importantnote:theseqnamemustbeoneusedwithinEnsembl,i.e.astandardchromosomenameoranEnsemblidentifiersuchasascaffoldID,withoutanyadditionalcontentsuchasspeciesorassembly.SeetheexampleGFFoutputbelow.2-source-nameoftheprogramthatgeneratedthisfeature,orthedatasource(databaseorprojectname)3-feature-featuretypename,e.g.Gene,Variation,4-start-Startpositionofthefeature,withsequencenumberingstartingat5-end-Endpositionofthefeature,withsequencenumberingstartingat6-score-Afloatingpoint7-strand-definedas+(forward)or-8-frame-Oneof'0','1'or'2'.'0'indicatesthatthefirstbaseofthefeatureisthefirstbaseofacodon,'1'thatthesecondbaseisthefirstbaseofacodon,andsoon..9-attribute-Asemicolon-separatedlistoftag-valuepairs,providingadditionalinformationabouteachfeature. GTF6612 gtfToGenePred-ignoreGroupsWithoutExonsGRCh38.gtfGRCh38.gtf. genePredToBedGRCh38.gtf. .predGRCh38.gtf.bed12 Read Read (.fastq(.fastq (2x150

Ensembl( Ensembl提供的參考組有2種組裝形式和3種重復(fù)序列處理方式,分別是primary,toplevel和unmasked(dna)、soft-masked(dna_sm)和masked(dna_rm)。一般選擇dna.primary或dna_sm.primary。Primaryassemblycontainsalltoplevelsequenceregionsexcludinghaplotypesandpatches.Thisfileisbestusedforperformingsequencesimilaritysearcheswherepatchandhaplotypesequenceswouldconfuseysis. Masked 組是指所有重復(fù)區(qū)和低復(fù)雜區(qū)被N代替，比對(duì)時(shí)就不會(huì)有 ，使得在允許錯(cuò)配的情況下，本來來自重復(fù)區(qū)的reads比對(duì)到組的其它位置。另外檢測(cè)重復(fù)區(qū)和低復(fù)雜區(qū)的軟件不可能是完美的，這就造成soft-masked組是指把所有重復(fù)區(qū)和低復(fù)雜區(qū)的序列用小寫字母標(biāo)出的組，由于主要的比對(duì)軟件，比如BWA、bowtie2等都忽略這些soft-組的比對(duì)效果和使用unmasked (2(2x150。 。SSTAR--runModegenomeGenerate--runThreadN2--genomeDirstar_GRCh38--genomeFastaFilesGRCh38.fa--sjdbGTFfileSSTAR--runModealignReads--readFilesIn (2x1503’readssoft

ReadSg Sg 的readscount和導(dǎo) 組瀏 SAM:SequenceAlignmentMapisatext-basedformatforstoringbiologicalsequencesalignedtoareferencesequence.SAM由頭文件和@@頭文件由一行行以@起始的注釋map結(jié)果,括reads 、起始位置 column1:gene

column3:countsforthe1streadstrandalignedwithRNA(htseq-countoption-syes)column4:countsforthe2ndreadstrandalignedwithRNA(htseq-countoption-sreverse) BAM:BinaryAlignment/Mapformat(compressedbinaryversionofSAMfile，節(jié)約存貯空間)。有的比對(duì)工具直接生成BAM格式比對(duì)結(jié)果（TopHat），有的比對(duì)工具生成SAM格式比對(duì)結(jié)果SAM轉(zhuǎn)BAM：samtoolsviewb-Sfile.samBAM文件查看：samtoolsviewfile.bam|BAM文件排序：samtoolssort-ofile.sorted.bamfile.bam(輸入文 samtoolsindexsamtoolsfaidx當(dāng)給出參考組的時(shí)候，會(huì)在第一排顯示參考組的序按下‘g’，則提示輸入要到達(dá)組的某一個(gè)位點(diǎn)。例子 ”表示到達(dá)第號(hào) 的第 個(gè)堿基位點(diǎn)處。（conda安裝的samtools,tivew功能用不了）

橫軸為樣本，縱軸為reads在組組比對(duì)特征的reads。右側(cè)表格與左側(cè)的 比對(duì)質(zhì)量評(píng)估–reads 轉(zhuǎn)錄組 reads理論上來源于exon（外顯子）區(qū)域。比對(duì)回 組后，這部分的reads應(yīng)該是最多