Length of reads in a RNAseq experiment

Question

[Carlos Caicedo]

Should all the reads of my experiment be of the same length if I am going to perform a de novo assembly?

The raw data contained reads of the same length but after the trimming process by quality some of them are reduced. I would like to know if there are some effect in the downstream analysis with this reads. (The majority of the reads remain of the same length and only a few reads have been trimmed)

Thanks

Carlos Andrés.

[Sinji]

Generally i've heard that you don't want to trim your RNA-seq reads as it could mess with downstream analysis (of course you'd remove adapter contamination).

[GouthamAtla]

It would be god to know what kind of trimming you did, like the program and command used.

[Carlos Caicedo]

I used Trimmomatic The “paired-end data” option was selected and the trimming steps were:

ILLUMINACLIP which cut adapter and other illumina-specific sequences from the read. The parameter for this step were:

• Truseq3 (Paired-ended, For miSeq and Hiseq)
• Maximum mismatch count which will still allow a full match to be performed: 2
• How accurate the match between the two 'adapter ligated' reads must be for PE palindrome read alignment: 30
• How accurate the match between any adapter etc. sequence must be against a read: 10

SLIDINGWINDOW which perform a sliding window trimming, cutting once the average quality within the window falls below a threshold. The parameter for this step were:

• Number of bases to average across: 4
• Average quality required: 20

Answer 1

[igor]

You should trim your reads before assembly. That's a perfectly reasonable assumption. I think any assembler will accept variable-length trimmed reads.