difference between 2 software of quality trimming

Question

[Aurélien Béliard]

hi,

I am trying to improve our DNA sequencing work-flow. I choose a software called AdapterRemoval for quality trimming and merging overlapping reads. I try to benchmark AdapterRemoval against the previous software called sickle with the same parameter and I don't understand why I have so many difference.

this is the sickle command with the minimum quality of 20 and minimum read length of 30 :

sickle pe -t illumina -f multiplex_bad_contam_1.fq -r multiplex_bad_contam_2.fq -q 20 -l 30 -o trimmed_multiplex_bad_contam_1_si.fq -p trimmed_multiplex_bad_contam_2_si.fq -s removed_sample_si.fq

This is the AdapterRemoval command the minimum quality and minimum length are the same:

AdapterRemoval --file1 multiplex_bad_contam_1.fq --file2 multiplex_bad_contam_2.fq --qualitybase 64  --trimqualities --minquality 20 --minlength 30 --output1 trimmed_multiplex_bad_contam_1_ar.fq --output2 trimmed_multiplex_bad_contam_2_ar.fq --singleton removed_sample_ar.fq

if someone have an idea it would be great.

thank for your help,

Aurel

ps: I apologise for the English mistakes but I am not naturally English speaker.

[Devon Ryan]

It's extremely likely that these tools work somewhat differently for determining exactly where to trim things (all trimmers works slightly differently), so I'm not surprised that you get different results with the same settings. You'll need to look through the code for each if the exact reasons are really important to you.

[Bioinformatics_NewComer]

It could be worth running post trimming process, and compare end results: length [mean, median, max, min], read/contigs, and other attributes from both tools. Just a thought. Different tools have different approach, same parameters may or may not yield same results.