Trimmomatic explanation

[tahamasoodi]

Can anybody explain the following command in Trimmonatic?

java -classpath trimmomatic-0.15.jar org.usadellab.trimmomatic.TrimmomaticPE s_1_1_sequence.txt.gz s_1_2_sequence.txt.gz lane1_forward_paired.fq.gz lane1_forward_unpaired.fq.gz lane1_reverse_paired.fq.gz lane1_reverse_unpaired.fq.gz ILLUMINACLIP:illuminaClipping.fa:2:40:15 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36

[DunderChief]

All of this is explained in detail on the trimmomatic website, but in general this will:

1. Clip Illumina adapters
2. then trim the leading nucleotides until quality > 3
3. then trim the trailing nucleotides until quality > 3
4. then using a sliding window of 4 nucleotides and trims when quality < 15
5. Remove any remaining sequences that are shorter than 36 nt.

You should really try looking at their documentation before posting questions here though.

[modi2020]

Hi,

I was wondering for steps 2 and 3, how many nucleotides would it trim from the start or end of the sequence. My intuition tells me its just one but I am not sure?

All of this is explained in detail on the trimmomatic website, but in general this will:

1. Clip Illumina adapters
2. then trim the leading nucleotides until quality > 3
3. then trim the trailing nucleotides until quality > 3
4. then using a sliding window of 4 nucleotides and trims when quality < 15
5. Remove any remaining sequences that are shorter than 36 nt.

You should really try looking at their documentation before posting questions here though.

[JackieBadger]

"trim the leading nucleotides until quality > 3"
Its pretty self explanatory...It will continue to trim bases that have a quality score lower than 3 until it hits a base where it is 3 or greater.

This it self is pretty useless as Ive never seen an illumina base with this low score and you would want to be retaining score at a very minimum of Q20

[mastal]

Trimmomatic explanation

"trim the leading nucleotides until quality > 3"
Its pretty self explanatory...It will continue to trim bases that have a quality score lower than 3 until it hits a base where it is 3 or greater.

This it self is pretty useless as Ive never seen an illumina base with this low score and you would want to be retaining score at a very minimum of Q20

It's useful for removing Ns (which have quality < 3) from the ends of the reads.

maria

[modi2020]

Thank you JackieBadger and mastal for your explaination.
I got the idea now.

[JackieBadger]

It's useful for removing Ns (which have quality < 3) from the ends of the reads.

maria

But you shouldn't be keeping any base that has a quality of between Q3-Q19
Wouldn't it just be better to trim off actual "N"s rather than assume they have are >Q3 ?

[mastal]

But you shouldn't be keeping any base that has a quality of between Q3-Q19
Wouldn't it just be better to trim off actual "N"s rather than assume they have are >Q3 ?

Yes, I agree, but I don't think you can do that with trimmomatic.

[modi2020]

Hi Jakie

I have tried to run trimmomatic with the option -phred20 to get reads with quality scores of Q20 but it doesn't seem to like it. It runs with -phred33 though. In their website, they say that it only accepts phred scores of 33 or 64. Am I wrong or is there any way of making it accept the -phred20 option?

Thank you

"trim the leading nucleotides until quality > 3"
Its pretty self explanatory...It will continue to trim bases that have a quality score lower than 3 until it hits a base where it is 3 or greater.

This it self is pretty useless as Ive never seen an illumina base with this low score and you would want to be retaining score at a very minimum of Q20

[JackieBadger]

Hi Jakie

I have tried to run trimmomatic with the option -phred20 to get reads with quality scores of Q20 but it doesn't seem to like it. It runs with -phred33 though. In their website, they say that it only accepts phred scores of 33 or 64. Am I wrong or is there any way of making it accept the -phred20 option?

Thank you

You are just a little confused.
Phred refers to the actual encoding of the quality score information http://en.wikipedia.org/wiki/FASTQ_format
....
phred33 or phred64 encryption are produced depend on the sequencer/software used to produce your data.

All quality codes have a range of quality scores associated with them and use different characters to ID particular quality scores (see wiki).

So you choose your quality encription (phredXX) and then choose the minimum quality score you want to enforce (e.g. 20)

[modi2020]

Thank you for the explanation Jakie. I got the idea now.
I tried to specify the quality after specifying the phred score and it didn't work though. To be specific I used -phred33 20
Is that what you meant?

You are just a little confused.
Phred refers to the actual encoding of the quality score information http://en.wikipedia.org/wiki/FASTQ_format
....
phred33 or phred64 encryption are produced depend on the sequencer/software used to produce your data.

All quality codes have a range of quality scores associated with them and use different characters to ID particular quality scores (see wiki).

So you choose your quality encription (phredXX) and then choose the minimum quality score you want to enforce (e.g. 20)

[modi2020]

Actually I think I got it.
The way I did it is using a sliding window option. What I think I did is ask it to go through the sequence in a window of 4 bps, take the average score, if the average score is below 20 then drop that window, otherwise keep moving. If the total length of the sequence after dropping low quality windows is less than 60 I removed it. I also used the leading and trailing options to drop low quality leading or trailing base pairs.
My complete command is as follows:

java -classpath trimmomatic-0.22.jar org.usadellab.trimmomatic.TrimmomaticPE -phred33 -trimlog trimmlog_log.txt R1.fastq R2.fastq Output_R1.fq unpaired_output1.fq Output_R2.fq unpairedoutput2.fq LEADING:20 TRAILING:20 SLIDINGWINDOW:4:20 MINLEN:60