Introducing the Trimmomatic

[tonybolger]

Hi All,

I've finally gotten around to making the Trimmomatic, a flexible read-trimming tool for Illumina NGS data available.

Instructions and download link are here. Any questions, requests etc, post 'em.

Since we still have issues getting clearance for including the illumina sequences (hopefully this will be resolved soon), the adapter trimming part won't be usable for now.

Enjoy and happy trimming.

[kga1978]

I have recently gone through a bunch of these programs and I gotta say this one is THE best by far - I'm only sorry I didn't find it until now.

Awesome tool - thanks for making it available!

[pmiguel]

Originally Posted by kga1978

I have recently gone through a bunch of these programs and I gotta say this one is THE best by far - I'm only sorry I didn't find it until now.

Awesome tool - thanks for making it available!

Which other tools did your try?

--
Phillip

[kga1978]

A bunch - some of them I can't remember as I uninstalled quickly. But I have tried cutadapt, far, btrim, SeqTrim, TagCleaner, prinseq, solexaQA - all nice tools but lacking in features and speed compared to this one (not sure why - but for one file, cutadapt took 3 hours, trimmomatic took 10 min - I almost thought it hadn't worked...).

[giorgifm]

Dear kg1978,

thanks for using Trimmomatic. We really did put effort in making it fully customizable, fast and memory efficient!

[kga1978]

Keep up the good work - it really is a good tool and so much faster than the others I have tried (still not sure why - but hey, who cares .

Only thing I have been missing - an output log showing how many reads were trimmed, what adapter, etc. That would be really useful.

[tonybolger]

Originally Posted by kga1978

Keep up the good work - it really is a good tool and so much faster than the others I have tried (still not sure why - but hey, who cares .

Only thing I have been missing - an output log showing how many reads were trimmed, what adapter, etc. That would be really useful.

Thanks for the complements, and glad trimmomatic is working well for you.

There is a trimlog (-trimlog <logfile>), but it currently only summarizes what happened to each read, not why. I agree an 'adapter tracking' log would also be useful.

[tonybolger]

A new version (0.2) is now available from the usual place

This version adds multi-threading, which makes it much much faster on multi-core machines, as well as correcting a bug in adapter detection code (which caused it to be overly-aggressive for around ~0.1% of the reads).

[neethav]

I am not able to run trimmomatic on MAC. It gives the error 'Failed to load Main-class manifest attribute from trimmomatic-0.20.jar'.
Is there a way to open this on MAC?
Thanks!

[tonybolger]

Originally Posted by neethav

I am not able to run trimmomatic on MAC. It gives the error 'Failed to load Main-class manifest attribute from trimmomatic-0.20.jar'.
Is there a way to open this on MAC?
Thanks!

I guess you've tried running it by double-clicking on the jar. Since trimmomatic doesn't have a GUI (yet), this won't work.

You need to run it from the terminal (Applications->Utilities->Terminal), using the command-line parameters as specified on the usadellab.org webpage.

[ETHANol]

Can someone help me understand the perimeters for ILLUMINACLIP.
I don't understand what the value for 'palindromeClipThreshold' or 'simpleClipThreshold' represent. In the example code they use values of 40 and 15, respectively but I have no idea what these numbers are doing. Also, does the value specified for 'palindromeClipThreshold' do anything in single end mode?

Thanks

ILLUMINACLIP:<fastaWithAdaptersEtc>:<seed mismatches>:<palindrome clip threshold>:<simple clip threshold>

fastaWithAdaptersEtc: specifies the path to a fasta file containing all the adapters, PCR sequences etc. The naming of the various sequences within this file determines how they are used. See below.

seedMismatches: specifies the maximum mismatch count which will still allow a full match to be performed

palindromeClipThreshold: specifies how accurate the match between the two 'adapter ligated' reads must be for PE palindrome read alignment.

simpleClipThreshold: specifies how accurate the match between any adapter etc. sequence must be against a read.

[tonybolger]

Originally Posted by ETHANol

Can someone help me understand the perimeters for ILLUMINACLIP.
I don't understand what the value for 'palindromeClipThreshold' or 'simpleClipThreshold' represent. In the example code they use values of 40 and 15, respectively but I have no idea what these numbers are doing. Also, does the value specified for 'palindromeClipThreshold' do anything in single end mode?

Sorry about the confusion.

These values are (very roughly speaking) log-10 probabilities of getting a match at random. Each perfectly matching base scores just over 0.6, so 15 requires a perfect 25 base match. Each mismatching base reduces the score by the Q/10 value of that base. So it takes 5 or even 6 additional matching bases to overcome one high quality mismatch, but maybe only 1 or 2 additional bases if the mismatching base is low quality.

"Palindrome" mode is only used in paired mode, when appropriately named "prefix" sequences are provided, and since it does 'end to end' matching of the pairs, it can afford a higher threshold, as suggested.

[ETHANol]

Ok, thanks!!! But now I have another question. So with a score of 15 there has to be 25 bp of adapter sequence for anything to get trimmed.

[tonybolger]

Originally Posted by ETHANol

Ok, thanks!!! But now I have another question. So with a score of 15 there has to be 25 bp of adapter sequence for anything to get trimmed.

Yep.

There's a trade-off:- you can set it lower, but then you will naturally be more likely to get something 'adapter-like' by random chance. I wouldn't recommend going below around 10 though.

[cllorens]

Hi Tony thank you for trimmomatic.

A question of newie about the manual.


When using the paired End Mode (below is as it is in your example of the manual)

java -classpath <path to trimmomatic jar> org.usadellab.trimmomatic.TrimmomaticPE [-threads <threads>] [-phred33 | -phred64] [-trimlog <logFile>] <input 1> <input 2> <paired output 1> <unpaired output 1> <paired output 2> <unpaired output 2> <step 1> ...


The question is which information is reported in the outputs "unpaired output 1" and "unpaired output 2" I installed the tool and I trying to make a couple of things with it but the files I have in return from the outputs are empty. Sure I am doing something wrong
but i could earn some time if you clarify me this doubt.

Best
Carlos

[cllorens]

Hi again,
Do not need answer for my question. I found this post where the author answer a similar question. Thank you anyway.

http://seqanswers.com/forums/showthread.php?t=10003

Carlos

[shawpa]

I am having trouble understanding the quality value used in trimmomatic. I want to trim bases from 3' and 5' ends that are below Q20. I have Illumina 1.8+ so my quality values are phred64. For the sample script they use 3 as the quality value. This can't be a Qscore of 3 so what is this?

[shawpa]

Originally Posted by shawpa

I am having trouble understanding the quality value used in trimmomatic. I want to trim bases from 3' and 5' ends that are below Q20. I have Illumina 1.8+ so my quality values are phred64. For the sample script they use 3 as the quality value. This can't be a Qscore of 3 so what is this?

Please disregard. I was being stupid. My values are in phred33.

[Niloofar]

Hi,

Would someone please tell me if there is a way to do trimming in Trimmomatic (using sliding window) on 3' end only?

Thanks

[westerman]

Take the example on the web page and only use the 'TRAILING:3 SLIDINGWINDOW:4:15' part of it. I.e., drop the 'ILLUMINACLIP:illuminaClipping.fa:2:40:15 LEADING:3 MINLEN:36' part of the example.