dslnc package is designed to retrieve known and novel long noncoding RNAs (lncRNAs) transcripts and analyze the disease-specific lncRNAs in multiple diseases through integrative analysis of RNA-seq data from an organ system. It also provides utilities for functional annotation of disease-specific lncRNAs. We believe that dslnc can accelerate the fundamental research in dissecting the functional role of lncRNAs in human diseases.
To run dslnc package, please install the following requirements.
- [FASTQ](Version 0.11.8)
- [HISAT2] (v2.0.5)
- [STRINGTIE] (v1.3.1 )
- [SAMTOOLS] (v0.1.19)
- [FeatureCounts] (v2.0.1)
- [Bedtools] (https://github.com/arq5x/bedtools2)
- [ISeeRNA] (http://www.myogenesisdb.org/iSeeRNA)
- [Rscript] (r-3.6.0)
# Disease1: https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos3/sra-pub-run-20/SRR8052751/SRR8052751.1
# Disease2: https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos2/sra-pub-run-15/SRR8052708/SRR8052708.1
cd dslnc/data
mkdir srr
wget https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos3/sra-pub-run-20/SRR8052751/SRR8052751.1
wget https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos2/sra-pub-run-15/SRR8052708/SRR8052708.1
# you can download sra-toolkits (https://hpc.nih.gov/apps/sratoolkit.html).
fastq-dump --split-files SRR8052751.1
fastq-dump --split-files SRR8052708.1Prepare the samples.lst file. The samples.lst file format should be prepared as follows:
CTRL_002_healthy ../data/srr/SRR8052751_1.fastq ../data/srr/SRR8052751_2.fastq
AD_004_lesional ../data/srr/SRR8052708_1.fastq ../data/srr/SRR8052708_2.fastqWrite the configure file, please refer to a template of the config.txt file. Please prepare config.txt file as follows:
# parameter_id<tab>path_to_directory or exec program
OUTDIR ./out
SAMPLE ./samples.lst
# parameter
READLEN 150
MINLEN 60
THREAD 24
# program setting
BIN bin
FASTQC fastqc
HISAT2 hisat2
STRINGTIE stringtie
SAMTOOLS samtools
Liftover liftOver (needed for hg38)
HOMER /homer/bin
# for STAR
GTF ../data/refGene.gtf or ../data/hg38.refGene.gtf (choose one version )
SPE human
INDEX ../data/hisat2_index/hg19 or ../data/hisat2_index/hg38 (choose one version )
CHROMSIZE ../data/hg19/hg19.chrom.size or ../data/hg38/hg38.chrom.size (choose one version )
# for dslnc analysis
Chr ../data/chr.lst
Bedtools bedtools
ISeeRNA iSeeRNA
Iseerna_conf ../data/iseerna.conf
FeatureCounts featureCounts
Rscript Rscript
Hg38ToHg19.over.chain ../data/hg38ToHg19.over.chain (needed for hg38)Create a makefile and run dslnc analysis automatically.
# Create a makefile
perl ../bin/hg19.dslnc.pl ../bin/config.txt (hg19)
# if analysis with hg38 as reference, please run as follows:
# perl ../bin/hg38.dslnc.pl ../bin/config.txt (hg38)
# run dslnc analysis
sh ../bin/auto_run.sh
Please make sure the directory looks like as follows:
├── out
│ ├── 1.standard
│ │ ├── CTRL_001_health
│ │ │ ├── 00datafilter
│ │ │ ├── 01alignment
│ │ │ ├── 02assembly
│ │ │ └── makefile
│ │ ├── makefile
│ │ └── PCA_002_lesional
│ │ ├── 00datafilter
│ │ ├── 01alignment
│ │ ├── 02assembly
│ │ └── makefile
│ ├── 2.identification
│ │ └── makefile
│ ├── 3.filter
│ │ └── makefile
│ └── 4.ds_score
│ └── makefile
└── samples.lstIf you would like to use example data for practicing the workflow, please download human RNAseq datasets from NCBI.
Disease1: https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos3/sra-pub-run-20/SRR8052751/SRR8052751.1
Disease2: https://sra-downloadb.be-md.ncbi.nlm.nih.gov/sos2/sra-pub-run-15/SRR8052708/SRR8052708.1Please prepare these files as mentioned in "Getting start" part.
# create makefile for RNA-seq analysis.
perl ../bin/dslnc.pl ../bin/config.txt
sh out/1.standard/standard.sh
# run RNA-seq analysis for all samples
cd out/1.standard && make && cd -In this step, it will run RNA-seq analysis in all samples one by one. It might take a long time in this step depending on your computer resource. You can obtain a transcriptome assembly for each sample.
├── out
│ ├── 1.standard
│ │ ├── CTRL_001_health
│ │ │ ├── 00datafilter
│ │ │ ├── 01alignment
│ │ │ ├── 02assembly
│ │ │ └── makefile
│ │ ├── makefile
│ │ └── PCA_002_lesional
│ │ ├── 00datafilter
│ │ ├── 01alignment
│ │ ├── 02assembly
│ │ └── makefile
│ ├── 2.identification
│ │ └── makefile
│ ├── 3.filter
│ │ └── makefile
│ └── 4.ds_score
│ └── makefile
└── samples.lstcd out/2.identification && make && cd -We combine all transcriptome assemblies and predict novel lncRNAs, please refer to test/out/2.identification/novo.lncrna.lst. This step also generate a merged gtf for all the samples, which can be found at test/out/2.identification/all.final.gtf.
├── 2.identification
│ ├── 00changebed.finished
│ ├── 00changegtf.finished
│ ├── 00getchr.finished
│ ├── 00rmknown.finished
│ ├── 00stringtie.finished
│ ├── 01getnovel.finished
│ ├── 01getnovelgtf.finished
│ ├── 01iseerna.finished
│ ├── 02changebed.finished
│ ├── 02getalllnclst.finished
│ ├── 02getfinalgtf.finished
│ ├── 02getgtf.finished
│ ├── 02getknownlnc.finished
│ ├── 02getknownlncgtf.finished
│ ├── 02getmergegtf.finished
│ ├── 02getnovellncgtf.finished
│ ├── 02rmIntergeniclnc.finished
│ ├── abinit.iseerna.novo.out
│ │ ├── 01make.finished
│ │ ├── abinit.merged.cl.novo.fa
│ │ ├── abinit.merged.cl.novo.feature
│ │ ├── abinit.merged.cl.novo.gtf
│ │ ├── abinit.merged.cl.novo.list
│ │ ├── abinit.merged.cl.novo.orf
│ │ ├── abinit.merged.cl.novo.predict
│ │ ├── abinit.merged.cl.novo.result
│ │ ├── abinit.merged.cl.novo.result.0.8.filtered
│ │ ├── abinit.merged.cl.novo.scaled
│ │ ├── abinit.merged.cl.novo.svm
│ │ ├── all_abinit.merged.cl.novo.consv
│ │ ├── consv
│ │ │ ├── chr10.consv
│ │ │ ├── chr10.dat
│ │ │ ├── chr10.list
│ │ │ ├── chr11.consv
│ │ │ ├── chr11.dat
│ │ │ ├── chr11.list
│ │ │ ├── chr12.consv
│ │ │ ├── chr12.dat
│ │ │ ├── chr12.list
│ │ │ ├── chr13.consv
│ │ │ ├── chr13.dat
│ │ │ ├── chr13.list
│ │ │ ├── chr14.consv
│ │ │ ├── chr14.dat
│ │ │ ├── chr14.list
│ │ │ ├── chr15.consv
│ │ │ ├── chr15.dat
│ │ │ ├── chr15.list
│ │ │ ├── chr16.consv
│ │ │ ├── chr16.dat
│ │ │ ├── chr16.list
│ │ │ ├── chr17.consv
│ │ │ ├── chr17.dat
│ │ │ ├── chr17.list
│ │ │ ├── chr18.consv
│ │ │ ├── chr18.dat
│ │ │ ├── chr18.list
│ │ │ ├── chr19.consv
│ │ │ ├── chr19.dat
│ │ │ ├── chr19.list
│ │ │ ├── chr1.consv
│ │ │ ├── chr1.dat
│ │ │ ├── chr1.list
│ │ │ ├── chr20.consv
│ │ │ ├── chr20.dat
│ │ │ ├── chr20.list
│ │ │ ├── chr21.consv
│ │ │ ├── chr21.dat
│ │ │ ├── chr21.list
│ │ │ ├── chr22.consv
│ │ │ ├── chr22.dat
│ │ │ ├── chr22.list
│ │ │ ├── chr2.consv
│ │ │ ├── chr2.dat
│ │ │ ├── chr2.list
│ │ │ ├── chr3.consv
│ │ │ ├── chr3.dat
│ │ │ ├── chr3.list
│ │ │ ├── chr4.consv
│ │ │ ├── chr4.dat
│ │ │ ├── chr4.list
│ │ │ ├── chr5.consv
│ │ │ ├── chr5.dat
│ │ │ ├── chr5.list
│ │ │ ├── chr6.consv
│ │ │ ├── chr6.dat
│ │ │ ├── chr6.list
│ │ │ ├── chr7.consv
│ │ │ ├── chr7.dat
│ │ │ ├── chr7.list
│ │ │ ├── chr8.consv
│ │ │ ├── chr8.dat
│ │ │ ├── chr8.list
│ │ │ ├── chr9.consv
│ │ │ ├── chr9.dat
│ │ │ ├── chr9.list
│ │ │ ├── chrM.consv
│ │ │ ├── chrM.dat
│ │ │ ├── chrM.list
│ │ │ ├── chrX.consv
│ │ │ ├── chrX.dat
│ │ │ ├── chrX.list
│ │ │ ├── chrY.consv
│ │ │ ├── chrY.dat
│ │ │ ├── chrY.list
│ │ │ └── Makefile
│ │ ├── iSeeRNA.conf
│ │ └── Makefile
│ ├── abinit.merged.cl.bed
│ ├── abinit.merged.cl.gtf
│ ├── abinit.merged.cl.novo.bed
│ ├── abinit.merged.cl.novo.gtf
│ ├── abinit.merged.gtf
│ ├── abinit.new.gtf
│ ├── all.final.gtf
│ ├── all.final.lncgeneid.lst
│ ├── all.lncRNA.gtf
│ ├── lncRNA.trans.gtf
│ ├── lncRNA.trans.lst
│ ├── makefile
│ ├── novo.rna.gtf
│ ├── novo.trans.bed
│ ├── novo.trans.gtf
│ ├── novo.trans.non-ol.bed
│ └── novo.trans.non-ol.gtfcd out/3.filter && make && cd -This step will quatify the expression level of all lncRNAs, please refer to ./test/out/3.filter/fpkm_result.txt. On the other hand, lncRNAs and protein coding genes (PCGs) with low expression levels will be filtered out (please refer to: /test/out/3.filter/all.lncRNA.filter.fpkm.txt and /test/out/3.filter/all.pcg.filter.fpkm.txt).
├── 3.filter
│ ├── 00featurecount.finished
│ ├── 01counttofpkm.finished
│ ├── 01getheader.finished
│ ├── 01getsampleid.finished
│ ├── 01rmheader.finished
│ ├── 02filter.finished
│ ├── 02getallfpkm.finished
│ ├── 02getalllncfpkm.finished
│ ├── 02lncfilter.finished
│ ├── 03getpcgfpkm.finished
│ ├── 03getpcglst.finished
│ ├── 03pcgfilterfpkm.finished
│ ├── all.lncRNA.filter.fpkm.txt
│ ├── all.lncRNA.fpkm.txt
│ ├── all.pcg.filter.fpkm.txt
│ ├── count.txt
│ ├── count.txt.summary
│ ├── fpkm_result.txt
│ ├── header.txt
│ ├── makefile
│ ├── pcg.fpkm.txt
│ └── refgene.codingene.lstcd out/4.ds_score && make && cd -The disease-specificity score of lncRNAs and protein coding genes is in test/out/4.ds_score/all.lncRNA.filter.dsscore.txt and test/out/4.ds_score/all.pcg.filter.dsscore.txt, respectively. destiny.pdf shows the comparion of DS score distribution between lncRNAs and protein coding genes.
└── 4.ds_score
├── 01getavg.finished
├── 01getlncavg.finished
├── 01lncds.finished
├── 02ds.finished
├── 02getpcgavg.finished
├── 02pcgds.finished
├── 03destiny.finished
├── 03getallscore.finished
├── 03getlncscore.finished
├── 03getpcgscore.finished
├── 03rmheader.finished
├── all.gene.filter.score.txt
├── all.lncRNA.filter.avg.fpkm.txt
├── all.lncRNA.filter.dsscore.txt
├── all.pcg.filter.avg.fpkm.txt
├── all.pcg.filter.dsscore.txt
├── codingscore.txt
├── destiny.pdf
├── makefile
├── noncodingscore.txt
└── Rplots.pdfThe format of disease-specifity score result.
#{id}\tab{disease1}\tab{disease2}\tab{DS_scor _of_gene}\tab{most_specific_disease}
LOC100133091 0.851892614542727 2.8187855877881 0.896895005858782 disease1
LOC101928126 3.02669783420868 1.16114685818554 0.86991908642307 disease2
RALY-AS1 1.06407345733899 1.59204120009034 0.773022319339166 disease1
AS-PTPRE 2.0092520953125 1.42828260748092 0.75973106098206 disease2This density plot shows DS score of lncRNAs (green) compared with PCGs (pink).
Liu et al. 2022 The long noncoding RNA atlas in healthy and diseased skin (manuscript in preparation).