15. Read Quantification

Learning objectives:

• Learn mapping and differential gene expression analysis of rna-seq data
• Interpret rna-seq analysis results

15.1. Make a new working directory and link the trimmed reads and assembly

We will be using the same trimmed fastq data as before (Tulin et al., 2013).

The following commands will create a new folder rnaseq and link the data in:

cd ..
mkdir -p rnaseq
cd rnaseq

ln -s ../trim/0Hour*.qc.fq.gz .
ln -s ../trim/6Hour*.qc.fq.gz .

ln -s /opt/rnaseq/assembly/nema_trinity/Trinity.fasta .

Note that this is a previously-assembly Trinity.fasta file that the whole class will use. Why is this important?

15.2. Index the assembly:

salmon index --index nema --type quasi --transcripts Trinity.fasta

15.3. Run salmon on all the samples:

• What are the flags used in the salmon command?
• Read up on libtype, here.

for R1 in *R1*.qc.fq.gz
do
  sample=$(basename $R1 extract.qc.fq.gz)
  echo sample is $sample, R1 is $R1
  R2=${R1/R1/R2}
  echo R2 is $R2
  salmon quant -i nema -p 2 -l IU -1 <(gunzip -c $R1) -2 <(gunzip -c $R2) -o ${sample}.quant
done

15.4. Take a look at quant output

• Examples of exploring the output

head 0Hour_ATCACG_L002_R1_001.qc.fq.gz.quant/quant.sf

15.5. Look at all the mapping rates:

find . -name \salmon_quant.log -exec grep -H "Mapping rate" {} \;

15.6. More reading

• Schurch et al. 2016. “How many biological replicates are needed in an RNA-seq experiment and which differential expression tool should you use?”.
• Patro et al. 2017 “Salmon provides accurate, fast, and bias-aware transcript expression estimates using dual-phase inference”.
• Salmon documentation