why i have less than 50% reads surviving in pair after trimming adapters?

[criscruz]

Hi All.

I'm using a ramdom amplification protocol and for library preparation I'm using Nextera XT. I'm performing seq step with Miseq reagent kit v3 600 cycles but i only use 200 cycles instead of 300 in order to get better quality score.

After adapters trimmed with Trimmomatic I've got this result:

TrimmomaticPE: Started with arguments:
-phred33 -trimlog 12143_trimlog.txt 12143_R1.fastq.gz 12143_R2.fastq.gz 12143_FWP.fq.gz 12143_FWUP.fq.gz 12143_REVP.fq.gz 12143_REVUP.fq.gz ILLUMINACLIP:/home/dell/miniconda3/share/trimmomatic/adapters/NexteraPE-PE.fa:2:30:10
Multiple cores found: Using 4 threads
Using PrefixPair: 'AGATGTGTATAAGAGACAG' and 'AGATGTGTATAAGAGACAG'
Using Long Clipping Sequence: 'GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG'
Using Long Clipping Sequence: 'TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG'
Using Long Clipping Sequence: 'CTGTCTCTTATACACATCTCCGAGCCCACGAGAC'
Using Long Clipping Sequence: 'CTGTCTCTTATACACATCTGACGCTGCCGACGA'
ILLUMINACLIP: Using 1 prefix pairs, 4 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences
Input Read Pairs: 2764988 Both Surviving: 658763 (23.83%) Forward Only Surviving: 2105922 (76.16%) Reverse Only Surviving: 288 (0.01%) Dropped: 15 (0.00%)
TrimmomaticPE: Completed successfully

FastQC basic statistics for forward before triimmig
Encoding Sanger / Illumina 1.9
Total Sequences 2764988
Sequences flagged as poor quality 0
Sequence length 201
%GC 42

FastQC basic statistics for forward paired after triimmig
Encoding Sanger / Illumina 1.9
Total Sequences 658763
Sequences flagged as poor quality 0
Sequence length 1-201
%GC 41
Why I have a very low % of both surviving and why a high value for forward only?

I really appreciate your feedbacks.


thanks

Chris

[kmcarr]

<CLIP>
Why I have a very low % of both surviving and why a high value for forward only?

I really appreciate your feedbacks.


thanks

Chris

This result indicates that your library included a large percentage of fragments with insert size less than or equal to 200bp in length. The way Trimmomatic paired end trimming operates (by default) when it finds the forward and reverse read overlap 100% (i.e. the insert is shorter than the read length) it keeps read 1 and discards read 2. The result is read 1 (forward) surviving but read 2 (reverse) not. When the overlap is only partial or not at all (i.e. insert is > 200bp) both read 1 and read 2 survive.

[criscruz]

thanks so much kmcarr