Question about trimmomatic

[janaahan13]

Hi,

I'm a first time user to trimmomatic and I have paired-end RNA seq data. My question why do I get paired and unpaired output for each read at the output?

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>

I have used CLC genomics for trimming and using that I'll get two output files namely read1_trimmed.fastq and read2_trimmed.fastq. Why do I get 4 output files in trimmmomatic?

[janaahan13]

I got the answer from Federico.

the answer is: because Trimmomatic doesn't discard unpaired reads, as CLC does.
Trimmomatic trims both reads in a pair. If one of the reads is of high quality, and the second one is of low quality, only the first is kept. This may happen in cases when the two sequencing cycle groups have generally different noise levels (recently, it happened in our institute that the laser died during the second cycle, and trimming removed most of the _2 reads, allowing us to keep the _1).

Most of the other non-CLC trimming tools do the same as Trimmomatic. If you really don't want them, you can ignore _U1 and _U2. Trimmomatic just give you an extra chance to use your unpaired sequences.

Federico

[Rahul shelke]

For assembly which reads should be used? paired one or unpaired...

[westerman]

For assembly which reads should be used? paired one or unpaired...

You can use both. Depends on the assembler.

[karl.stefic]

Hi everybody,

Just joined the forum, hope you're still following the thread.

To keep on the subject, is it possible to assemble both datasets (paired+unpaired) in order to perform an assembly (Trinity) and not lose precious data.

Thanks