salmon使用体验

salmon转录本定量

brief

第一点是，通常说的转录组分析其中有一项是转录本定量，这是一个很trick的说话，说成定量/quantify要适合一些。
因为我们可以根据 reads summary的方式分为两种定量，一种是 transcript-level quantify,一种是 gene-level quantify。

第二点 transcript-level quantify根据比对方式又可以细分：

Transcript quantification 大致分为两类：

• Alignment-based transcript quantification
  这里的比对也要作出区分，一种是和基因组比对（aligns reads to the reference genome），一种是和转录组比对（aligns reads to the reference transcriptome）。
  和基因组比对后，可以利用Cufflinks or StringTie 等tools，不仅可以测算已知转录本丰度还可以发现新的转录本
  和转录组比对后，可以利用RSEM or eXpress or Salmon-Aln 等tools进行转录本丰度的测算
• Alignment-free transcript quantification
  就是直接 assign reads directly to transcripts，比如ailfish, Salmon-SMEM,quasi-mapping, and kallisto 这些工具可以实现。

---

conda activate NGS
conda config --add channels bioconda

conda search salmon
conda install salmon

salmon --help

salmon v0.14.1

Usage: salmon -h|–help or
salmon -v|–version or
salmon -c|–cite or
salmon [–no-version-check] [-h | options]

Commands:
index Create a salmon index
quant Quantify a sample
alevin single cell analysis
swim Perform super-secret operation
quantmerge Merge multiple quantifications into a single file

模式一：fastq作为输入文件

# step 1
# make salmon map index
# The index is a structure that salmon uses to quasi-map RNA-seq reads during quantification.
wget https://ftp.ensembl.org/pub/release-111/fasta/macaca_mulatta/cdna/Macaca_mulatta.Mmul_10.cdna.all.fa.gz

# 软件作者希望制作 decoy awere transcriptome，我没管，直接运行下下面的口令
salmon index -t ./Macaca_fascicularis.Macaca_fascicularis_6.0.cdna.all.fa.gz  -i ./Macaca_fascicularis.Macaca_fascicularis_6.0.cdna.all.salmon.index

ls Macaca_fascicularis.Macaca_fascicularis_6.0.cdna.all.salmon.index/
# duplicate_clusters.tsv  hash.bin  header.json  indexing.log  quasi_index.log  refInfo.json  rsd.bin  sa.bin  txpInfo.bin  versionInfo.json
# make index done 

# step 2
# rawdata QC
datadir="/public/Project_datasets/HY0007/Macaca_fascicularis_RNA-seq/PM-XS05KF2023080038-05四川大学6个食蟹猴普通真核有参转录组建库测序分析任务单/ANNO_XS05KF2023080038_PM-XS05KF2023080038-05_2023-11-02_16-01-18_22C2TNLT3/Rawdata"
mkdir 20240426-NGS-6samples
cd 20240426-NGS-6samples/

mkdir qc
ls ${datadir}/ |while read id ;do echo ${datadir}/${id}/${id}_R1.fq.gz;done > qc.txt
ls ${datadir}/ |while read id ;do echo ${datadir}/${id}/${id}_R2.fq.gz;done >> qc.txt

cat qc.txt |while read id ;do (fastqc -o ./qc $id &);done
multiqc ./qc -o ./multiqc

# trim-adaptor
###
cat qc.txt |grep "R1" >1.txt
cat qc.txt |grep "R2" >2.txt
paste 1.txt 2.txt  > trim.txt
cat trim.txt |while read id;do a=($id) && echo /public/home/djs/software/TrimGalore-master/trim_galore -q 25 --phred33 --stringency 3 -o ./Clean_data --paired ${a[0]} ${a[1]}; done > parafly.txt

# conda install -c bioconda parafly
nohup ParaFly -c parafly.txt -CPU 15 >>out.log 2>>err.log &

###
cd Clean_data
mkdir {Fastqc,Multiqc}

ls |grep "val.*fq" |while read id ;do (fastqc -o ./Fastqc ./$id &);done
multiqc ./Fastqc -o ./Multiqc

# step3 map to reference
index="/public/home/djs/reference/macaca_fascicularis/Macaca_fascicularis.Macaca_fascicularis_6.0.cdna.all.salmon.index/"
# 双端测序
cd /public/home/djs/huiyu/project/HY0007/20240426-NGS-6samples/Clean_data

ls *val*gz|cut -d"_" -f 1|sort -u |while read id;do
echo salmon quant -i $index -l ISF --gcBias \
 -1 ${id}_R1_val_1.fq.gz -2 ${id}_R2_val_2.fq.gz -p 2 \
 -o ../salmon_output/${id}_output 
done > paired_salmon.sh

nohup ParaFly -c paired_salmon.sh -CPU 12 >>out.log 2>>err.log &

# quantify result
ls
# aux_info  cmd_info.json  lib_format_counts.json  libParams  logs  quant.sf
head quant.sf
# ENSMFAT00000064841.2    354     110.000 37.761624       31.000
# ENSMFAT00000064566.2    372     126.000 107.406974      101.000
# ENSMFAT00000061855.2    336     96.000  108.422787      77.680
# ENSMFAT00000061921.2    336     96.000  67.838005       48.603
# ENSMFAT00000061935.2    336     96.000  185.240776      132.717
# ENSMFAT00000097936.1    252     45.460  20.632500       7.000
# ENSMFAT00000064576.2    348     107.929 130.356210      105.000
# ENSMFAT00000064726.2    339     98.000  60.160059       44.000
# ENSMFAT00000064735.2    267     52.000  10.307143       4.000

## 进入R工作
# 进入R做一下 FPKM得转换countToTpm <- function(counts, effLen)
countToTpm <-  function(counts, effLen){
    rate <- log(counts) - log(effLen)
    denom <- log(sum(exp(rate)))
    exp(rate - denom + log(1e6))
}
 
countToFpkm <- function(counts, effLen){
    N <- sum(counts)
    exp( log(counts) + log(1e9) - log(effLen) - log(N) )
}
 
fpkmToTpm <- function(fpkm){
    exp(log(fpkm) - log(sum(fpkm)) + log(1e6))
}

df <- read.table("quant.sf",header=T)
# 过滤低表达基因
df <- df[df$NumReads >=10,]
# normalization
df$fpkm <- countToFpkm(df$NumReads,df$EffectiveLength)
df$tpm <- countToTpm(df$NumReads,df$EffectiveLength)
write.csv(df,file="quant_filter_transform.csv",row.names=F)

files <- list.files()
dlist <- lapply(files,function(file){read.table(paste("./",file,"/quant_filter_transform.csv",sep=""),header=T,sep=",")})

需要特别注意得地方

• 参数 —libType A 的设置，一般情况是 --libType ISF 或者 --libType A 让软件自己推测。
  

模式二： bam文件作为输入

这个bam文件是fastq文件与参考转录组比对的结果，注意不是与参考基因组的比对结果。
然后 transcripts.fa 是参考转录组文件（这种模式下，可以不用建议参考转录组的index）。

> ./bin/salmon quant -t transcripts.fa -l <LIBTYPE> -a aln.bam -o salmon_quant
# quantify result
ls
# aux_info  cmd_info.json  lib_format_counts.json  libParams  logs  quant.sf
head quant.sf
# ENSMFAT00000064841.2    354     110.000 37.761624       31.000
# ENSMFAT00000064566.2    372     126.000 107.406974      101.000
# ENSMFAT00000061855.2    336     96.000  108.422787      77.680
# ENSMFAT00000061921.2    336     96.000  67.838005       48.603
# ENSMFAT00000061935.2    336     96.000  185.240776      132.717
# ENSMFAT00000097936.1    252     45.460  20.632500       7.000
# ENSMFAT00000064576.2    348     107.929 130.356210      105.000
# ENSMFAT00000064726.2    339     98.000  60.160059       44.000
# ENSMFAT00000064735.2    267     52.000  10.307143       4.000