5. Trimming

We will use Trimmomatic for read trimming and filtering, and adapters removal.

Let’s see its general usage. To run trimmomatic you need to specify its full path: /cm/shared/apps/trimmomatic/0.33/trimmomatic-0.33.jar):

$ java -jar /cm/shared/apps/trimmomatic/0.33/trimmomatic-0.33.jar -h

Usage:

 PE [-threads <threads>] [-phred33|-phred64] [-trimlog <trimLogFile>] [-basein <inputBase> | <inputFile1> <inputFile2>] 
      [-baseout <outputBase> | <outputFile1P> <outputFile1U><outputFile2P> <outputFile2U>] <trimmer1>...

or:

SE [-threads <threads>] [-phred33|-phred64] [-trimlog <trimLogFile>] <inputFile> <outputFile> <trimmer1>...

For paired end reads (PE), after specifying some options, the arguments are the input forward and reverse fastq files, the names of the paired and unpaired remaining reads after trimming for the forward reads, and the names of the paired and unpaired remaining reads after trimming for the reverse reads. Finally, you need to specify the parameters for the trimming operations. Since the Trimmomatic output files are fastq files, you must be careful of not mixing them with the original un-trimmed fastq files, or to overwrite the un-trimmed fastq files. You can create a specific folder for the trimmed files, or use specific names that will remind you what they are. Let’s try for the forward 2-cells fastq file. We will use again the submission script but with larger memory limits (lets say 40GB. The java VM consumes a lot of memory) and the running command for trimmomatic:

#$ -S /bin/sh
#$ -cwd
#$ -q amd.q,large.q,intel.q
#$ -l h_vmem=40G
#$ -e run_trimmomatic.e
#$ -N run_trimmomatic
#$ -o run_trimmomatic.o
java -jar /cm/shared/apps/trimmomatic/0.33/trimmomatic-0.33.jar PE -threads 2 -phred33 data/2cells_1.fastq data/2cells_2.fastq data/2cells_1.trim.fastq data/2cells_1.trim.unpaired.fastq data/2cells_2.trim.fastq data/2cells_2.trim.unpaired.fastq LEADING:20 TRAILING:20 AVGQUAL:20 MINLEN:25