Trimmomatic: Higher number of "forward only" than "reverse only" surviving reads

Question

[Lucila]

Hi all, after running Trimmomatic (Illumina, 150bp paired end reads), I found that I have a lot more forward only surviving reads than reverse only surviving reads (7-8% vs. 0.5% approximately) for all my samples. Is this normal? Any ideas about why does it happen?

Thank you in advance for your help, Lucila.

[h.mon]

What is the command you are using? Did you check the quality of the reads before trimming? It is not uncommon for reverse reads to have slight lower quality than forward reads, so quality trimming would affect R2 more.

[Lucila]

Thank you for your answer. Here is an example of the command that I used:

java -jar trimmomatic-0.36.jar PE -threads 20 -phred33 C1_1.fq.gz C1_2.fq.gz trimmed/C1.paired.1.fastq.gz trimmed/C1.single.1.fastq.gz trimmed/C1.paired.2.fastq.gz trimmed/C1.single.2.fastq.gz TRAILING:5 LEADING:5 SLIDINGWINDOW:4:15 ILLUMINACLIP:TruSeq3-PE-2.fa:2:30:10 MINLEN:50 2> C1.trim.log

I have checked the quality of all my samples before trimming and the parameters Per base sequence quality and Per sequence quality scores were OK for all the samples. I obtained FAIL in the following parameters: Per base sequence content, Per base GC content, and Sequence Duplication Levels.

[GenoMax]

I obtained FAIL in the following parameters: Per base sequence content, Per base GC content, and Sequence Duplication Levels.

FAIL needs to be taken with a big pinch salt (I assume you are referring to FastQC results). You should read informative blog posts by group that wrote FastQC here. Most probably there is no problem with your sequence data. There is likely no need to do TRAILING/LEADING trimming either.

[Lucila]

Thank you so much genomax! Very useful link.

[mastal511]

See this previouis discussion about Trimmomatic. A: Interpretation of Trimmomatic Results after Paired-End Adapter Trimming

By default, trimmomatic deletes the reverse read when it trims adapters from paired reads, leaving an excess of unpaired forward reads.

[GenoMax]

Just to be sure, you did trim them together correct?

[Lucila]

I put both libraries in the same command, I have copied an example of the command I have used in the previous answer.

[gtrwst9]

(deleted by the author)

[nidhi.vijayan13]

I am having the same issue. But when I align to the genome using bowtie2, I get an error:

Error, fewer reads in file specified with -2 than in file specified with -1 terminate called after throwing an instance of 'int' (ERR): bowtie2-align died with signal 6 (ABRT)

[GenoMax]

You can try repair.sh from BBMap suite to re-sync paired end reads in the two files and remove singletons. This happens when you scan/trim paired end reads independently.