Scapture

SCAPTUREDescription:SCAPRUTE pipeline. A deep learning-embedded pipeline that captures polyadenylation information from 3 prime tag-based RNA-seq of single cellsInstallation:Require Bedtools v2.26.0Require featureCounts ≥ v1.6.2Create conda env from SCAPTURE yaml file: conda env create -f SCAPTURE_env.yamlDownload and unzip package from Github, then add "+x" to scripts to make it executable. unzip SCAPTURE-main.zip cd SCAPTURE-main/ chmod +x scapture scapture_annotation scapture_callpeak scapture_evaluate scapture_mergesample scapture_quantExport path of SCAPTURE suite to environment SCAPTURE="/Your_Path/SCAPTURE-main"Add SCAPTURE to ./bashrc file of your environment.> ~/.bashrc"> echo "export ${SCAPTURE}:\$PATH" >> ~/.bashrcOr manually export to environment variable before running SCAPTURE export PATH=${SCAPTURE}:$PATHTest whether SCAPTURE could be found which scapture "/Your_Path/SCAPTURE-main/scapture"Activate SCAPTURE_env environment before running SCAPTURE conda activate SCAPTURE_envUsage: [options] -h/---help -- help information">scapture -m [options] -h/---help -- help informationModule: annotationOptions:-o -- prefix of the output files-g -- genome .fa file--gtf -- gene annotation file (GTF format, recomannd GENOCODE annotation with "gene_name" and "gene_type" tags)--cs -- chromosom size file--extend -- extended dowstream distance of 3 prime gene annotation to call peaks (bp, default: 2000)-h/---help -- help informationModule: PAScallOptions:-a -- prefix of output annotation files in "annotation" module-o -- prefix of output file-b -- Cell Ranger aligned BAM file-g -- genome .fa file-p -- number of cores-l -- read length of sequenceing cDNA in scRNA-seq-w -- width of poly(A) peaks (default: 400)--species -- species for DeepPASS model ('human', 'mouse')--overlap -- overlapped ratio of exonic peaks to merge (default: 0.5)--polyaDB -- poly(A) site database bed6 file (optional)-h/---help -- help informationModule: PASmergeOptions:-o -- prefix of output file--peak -- list of evaluated peak files to merge (one sample per line, split by tab, 1st col: "Sample_name", 2cd col: "PathofEvaluatedPeakFile" )--path -- path of scapture suite (ignore if scapture in PATH)--rawpeak -- raw peak files to merge (Restricted by --peak)-h/---help -- help informationModule: PASquantOptions:-o -- output prefix-p -- number of threads-b -- Cell Ranger aligned BAM file--pas --

PASs peaks file to quantify--celllist -- cell barcode file as white list(one barcode per line), or use "--celllist FromBAM" to used all cell barcode from input BAM (for unfiltered BAM, it might take lots of time).--celltype -- cell type annotation file (tab split, 1st col: "cell_barcode", 2cd col: "cell_type" ).--bw -- generate bigwiggle file for each cell type in --celltype option (true, false, default: false)-h/---help -- help informationVersion: 1.0 2021/01/25Author: Guo-Wei Li Email: liguowei@picb.ac.cnStep-by-step protocolWe use six human PBMC scRNA-seq datasets from 10x Genomics to show step-by-step protocol of SCAPTURE. All files used/generated by this example could be found at shared Google Drive.File preparationMake sure that featureCounts and SCAPTURE are available in your PATH. User can manually add software path to PATH, like: export PATH=/home/software/SCAPTURE-main:$PATH export PATH=/home/software/subread-1.6.2-Linux-x86_64/bin:$PATHPrepare gene annotation and genome file. In this example, we use annotation from GENCODE hg38. SCAPTURE supports references given by users for specific study. Currently, we take GENCODE GTF file as gene annotation. GENCODE GTF annotation contains "gene_name" and "gene_type" in the 8th column field, and SCAPTURE utilizes the information to annotate PAS from genes. However, it is not necessary for the future (we are working on it). gencode.v34.annotation.gtf hg38.chromsize genome.fa genome.fa.faiIf poly(A) databases file (BED format, optional) is given, SCAPTURE will annotate PAS with database information and select relatively high confident PAS during merging different samples (in SCAPTURE PASmerge step). Combined PolyADB3, PolyA-Seq, PolyASite and GENCODE poly(A) databases file: human hg38 (SupTab_KnownPASs_fourDBs.txt), mouse mm10 (mm10.SupTab_KnownPASs_fourDBs.txt). SupTab_KnownPASs_fourDBs.txtFor running demo, we extracted reads in chr1:1-10000000 region from aligned BAM file of PBMCs in out study. And the white cell barcode list in each sample (generated by Cell Ranger count or given by the user). All used files could be found at here. ls ./PBMC* PBMC10k.test.bam PBMC8k.test.bam PBMC6k.test.bam PBMC5k.test.bam PBMC4k.test.bam PBMC3k.test.bam PBMC10k.test.bam.bai PBMC8k.test.bam.bai PBMC6k.test.bam.bai PBMC5k.test.bam.bai PBMC4k.test.bam.bai

PBMC3k.test.bam.bai PBMC10k.celllist PBMC8k.celllist PBMC6k.celllist PBMC5k.celllist PBMC4k.celllist PBMC3k.celllistRun scapture annotation moduleIn this step, SCAPTURE will generate annotation files for downstream analysis. If your study includes multiple samples, we recommend using the same annotation files in this step to analyze all samples. annotation.log">scapture -m annotation -o SCAPTURE_annotation -g genome.fa --gtf hg38.gtf --cs hg38.chromsize --extend 2000 &> annotation.logOutput files in this step:ls SCAPTURE_annotation*SCAPTURE_annotation.chromsize SCAPTURE_annotation.genetype.bed SCAPTURE_annotation.genetype.gtfSCAPTURE_annotation.element.txt SCAPTURE_annotation.genetype.extended.bed SCAPTURE_annotation.genetype.txtRun scapture PAScall moduleSCAPTURE takes aligned BAM files and above annotation files as input to call peaks. The PAScall module contains several steps:(1) perform transcript level peak calling and filtering(2) assign peak to APA-transcript (see detail in our paper)(3) evaluate peak with DeepPASS prediction (automatically run) and poly(A) database annotation (optional) PBMC3k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC4k.test.bam -l 98 -o PBMC4k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC4k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC5k.test.bam -l 91 -o PBMC5k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC5k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC6k.test.bam -l 98 -o PBMC6k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC6k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC8k.test.bam -l 98 -o PBMC8k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC8k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC10k.test.bam -l 91 -o PBMC10k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC10k.PAScall.log">scapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC3k.test.bam -l 98 -o PBMC3k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC3k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC4k.test.bam -l 98 -o PBMC4k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC4k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC5k.test.bam -l 91 -o PBMC5k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC5k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC6k.test.bam -l 98 -o PBMC6k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC6k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g

Genome.fa -b PBMC8k.test.bam -l 98 -o PBMC8k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC8k.PAScall.logscapture -m PAScall -a SCAPTURE_annotation -g genome.fa -b PBMC10k.test.bam -l 91 -o PBMC10k -p 16 --species human --polyaDB SupTab_KnownPASs_fourDBs.txt &> PBMC10k.PAScall.logOutput files in this step:#for the PBMC3k sample, we got:PBMC3k.exonic.peaks.bed # filtered exonic peaksPBMC3k.exonic.peaks.annotated.bed # assigned exonic peaksPBMC3k.exonic.peaks.evaluated.bed # evaluated exonic peaksPBMC3k.intronic.peaks.bed # filtered intronic peaksPBMC3k.intronic.peaks.annotated.bed # assigned intronic peaksPBMC3k.intronic.peaks.evaluated.bed # evaluated intronic peaksPBMC3k.3primeExtended.peaks.annotated.bed # assigned peaks in dowstream region beyond 3' end of genesPBMC3k.3primeExtended.peaks.evaluated.bed # evaluated peaks in dowstream region beyond 3' end of genesThe output poly(A) site file was in widely-used BED format see.Briefly,the 1-12 columns represent the spliced peak region of PAS. seeSpecifically, for "evaluated.bed",the 13th column is "number of refernce poly(A) sites supporting the PAS."the 14th column is "DeepPASS prediction of the PAS."the 15th column is "±100 nt sequence around the cleavage site of PAS."Run scapture PASmerge moduleIf your study includes multiple samples, SCAPTURE can merge the peak result from all samples, and create a combined peak reference file for downstream analysis. PBMC_ALL.exonic.peaklistfor i in 3 4 5 6 8 10; do echo -e PBMC${i}k"\t"PBMC${i}k.intronic.peaks.evaluated.bed; done > PBMC_ALL.intronic.peaklistfor i in 3 4 5 6 8 10; do echo -e PBMC${i}k"\t"PBMC${i}k.3primeExtended.peaks.evaluated.bed; done > PBMC_ALL.3primeExtended.peaklist#run the PASmerge modulescapture -m PASmerge -o PBMC_ALL.exonic --peak PBMC_ALL.exonic.peaklist &> PBMC_ALL.PASmerge.exonic.logscapture -m PASmerge -o PBMC_ALL.intronic --peak PBMC_ALL.intronic.peaklist &> PBMC_ALL.PASmerge.intronic.logscapture -m PASmerge -o PBMC_ALL.3primeExtended --peak PBMC_ALL.3primeExtended.peaklist &> PBMC_ALL.PASmerge.3primeExtended.log">#create table with sample name and sample peak file path:for i in 3 4 5 6 8 10; do echo -e PBMC${i}k"\t"PBMC${i}k.exonic.peaks.evaluated.bed; done > PBMC_ALL.exonic.peaklistfor i in 3 4 5 6 8 10; do echo -e PBMC${i}k"\t"PBMC${i}k.intronic.peaks.evaluated.bed; done > PBMC_ALL.intronic.peaklistfor i in 3 4 5 6 8 10; do echo -e PBMC${i}k"\t"PBMC${i}k.3primeExtended.peaks.evaluated.bed; done > PBMC_ALL.3primeExtended.peaklist#run the PASmerge modulescapture -m PASmerge -o PBMC_ALL.exonic --peak PBMC_ALL.exonic.peaklist &> PBMC_ALL.PASmerge.exonic.logscapture -m PASmerge -o PBMC_ALL.intronic --peak PBMC_ALL.intronic.peaklist &>

This instead of the 'Go To' button used in former releases. Premiere Pro default keyboard shortcuts Many commands have keyboard shortcut equivalents, so you can complete tasks with minimal use of the mouse. You can also create or edit keyboard shortcuts. File CommandsWindowsmacOSProject...Ctrl + Alt + NOpt + Cmd + NSequence...Ctrl + NCmd + NBinCtrl + /Cmd + /Open Project...Ctrl + OCmd + OClose ProjectCtrl + Shift + WShift + Cmd + WCloseCtrl + WCmd + WSaveCtrl + SCmd + SSave As...Ctrl + Shift + SShift + Cmd + SSave a Copy...Ctrl + Alt + SOpt + Cmd + SCapture...F5F5Batch Capture...F6F6Import from Media BrowserCtrl + Alt + IOpt + Cmd + IImport...Ctrl + ICmd + IExport MediaCtrl + MCmd + MGet Properties for Selection...Ctrl + Shift + HShift + Cmd + HExitCtrl + QCmd + Q Edit CommandsWindowsmacOSUndoCtrl + ZCmd + ZRedoCtrl + Shift + ZShift + Cmd + ZCutCtrl + XCmd + XCopyCtrl + CCmd + CPasteCtrl + VCmd + VPaste InsertCtrl + Shift + VShift + Cmd + VPaste AttributesCtrl + Alt + VOpt + Cmd + VClearDeleteForward DeleteRipple DeleteShift + DeleteShift + Forward DeleteDuplicateCtrl + Shift + /Shift + Cmd + /Select AllCtrl + ACmd + ADeselect AllCtrl + Shift + AShift + Cmd + AFind...Ctrl + FCmd + FEdit OriginalCtrl + ECmd + EKeyboard ShortcutsCtrl + Alt + KCmd + Opt + K Clip CommandsWindowsmacOSMake Subclip...Ctrl + UCmd + UAudio Channels...Shift + GShift + GAudio GainGGSpeed/Duration...Ctrl + RCmd + RInsert,,Overwrite..EnableShift + EShift + Cmd + ELinkCtrl + LCmd + LGroupCtrl + GCmd + GUngroupCtrl + Shift + GShift + Cmd + G Sequence CommandsWindowsmacOSRender Effects in Work AreaEnterEnterMatch FrameFFReverse Match FrameShift + RShift + RAdd EditCtrl + KCmd + KAdd Edit to All TracksCtrl + Shift + KShift + Cmd + KTrim EditShift + TCmd +