Trimmomatic no output files

[jgregg]

I'm new to Trimmomatic and am having trouble producing the 4 output files from a PE trim. I am running Trimmomatic 0.32 in OSX. When I issue the following command in terminal:

jlg:~ jgregg$ java -jar /Applications/Trimmomatic/Trimmomatic-0.32/trimmomatic-0.32.jar PE -threads 2 -phred33 -trimlog /Users/jgregg/Documents/ICH_Genomics/IchPE76/PE76_filtered_reads/Process_Testing/TrimLog235 /Users/jgregg/Documents/ICH_Genomics/IchPE76/PE76_filtered_reads/Process_Testing/Test_R1.fq.gz /Users/jgregg/Documents/ICH_Genomics/IchPE76/PE76_filtered_reads/Process_Testing/Test_R2.fq.gz -baseout /Users/jgregg/Documents/ICH_Genomics/IchPE76/PE76_filtered_reads/Process_Testing/TestOUT_IA235.fq.gz ILLUMINACLIP: /Users/jgregg/Documents/ICH_Genomics/IchPE76/PE76_filtered_reads/Process_Testing/Trimmo_IA235.fa:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:37

This runs for a few minutes and then I get:

Input Read Pairs: 17222594 Both Surviving: 13577333 (78.83%) Forward Only Surviving: 3273911 (19.01%) Reverse Only Surviving: 149140 (0.87%) Dropped: 222210 (1.29%)

This looks right, but the only files that are produced are the TrimLog file, and two files (TestOUT_IA235.fq.gz and Trimmo_IA235.fa/2/30/10) that contain sequences that have adapters. These appear to be the half of the sequence pairs that should have been discareded. Any help that people can offer would be greatly appreciated.

Jake

[mastal]

Your command looks correct, and the output that trimmomatic is reporting looks sort of right, but I have not run Trimmomatic on OSX or used v.032.

Have you tried running it without -baseout, and just specifying the 4 output file names?

Does trimmomatic correctly recognise your parameters, what output do you get when trimmomatic starts running?

[Birdman]

As mastal suggested, I would try specifying output files, after your inputs, e.g.

trimmomatic-0.32.jar PE <options> .../Test_R1.fq.gz .../Test_R1.fq.gz R1.paired.fq R2.paired.fq R1.unpaired.gq R2.unpaired.fq ...

[mastal]

the usual order of specifying the 4 output files for trimmomatic is:

R1_P.fq R1_U.fq R2_P.fq R2_U.fq

[Birdman]

You're right, my mistake.

[tonybolger]

Hi,

This one is kinda embarrassing - the -baseout flag actually doesn't work the way the help suggests it should in all cases.

The 'option' arguments i.e those with a "-" must currently come before the non-option arguments, such as the old style list of files for input/output and the trimming steps.

Also, you have a space between the ILLUMINACLIP and the adapter file name, which will also cause problems.

Tony.

[jgregg]

Thanks all,

Specifying the files and removing space got everything working, and it all looks right.

I do have one other question. When specifying the /2 sequence for palindrome mode, should you use the reverse complement of the expected adapter. I ask because FastQC presents the reverse compliment as an over-represented sequence in my R2 reads.

Thanks Again,

Jake

[tonybolger]

Thanks all,

Specifying the files and removing space got everything working, and it all looks right.

OK. I'll get a fix out for this as soon as i can spare some time.

I do have one other question. When specifying the /2 sequence for palindrome mode, should you use the reverse complement of the expected adapter. I ask because FastQC presents the reverse compliment as an over-represented sequence in my R2 reads.

With read-through (due to short insert fragments), the forward adapter shows up in reverse complement in the reverse read, and vice versa. However, trimmomatic still expects it to be named as the forward adapter (with the additional T included, which matches the A added during adapter ligation). In any case, unless you have special library preps, one of the included adapter files should do the job.

Tony.

[jgregg]

Tony,

Thanks so much, and thanks for Trimmomatic. I've read several post tonight and don't envy you, explaining the palindrome trimming. I think that I have it figured out now. Running some test to be sure.

Thanks again.

Jake