can't remove subset of adapter sequences

[jasbhead]

FastQC with raw reads indicated contamination by adapter sequences at the 3' end. Running IlluminaClip from Trimmomatic seemed to help the issue but adapter sequences appear to remain at the 3' end of at the 125-150bp positions. Any advice on what is going on and how I can go about troubleshooting? Attached are the FastQC Kmer enrichment plots for the same read file, untrimmed, trimmed and trimmed with clip threshold lowered from 30 to 20 (palindrome setting).

TruSeq Universal Adapter
5’ AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT

TruSeq Adapter, Index 6
5’ GATCGGAAGAGCACACGTCTGAACTCCAGTCACGCCAATATCTCGTATGCCGTCTTCTGCTTG

[mastal]

Are you only doing palindrome clipping?

How long are the adapter sequences for palindrome clipping in your IlluminaClipping.fasta file?

I find that sometimes palindrome clipping alone doesn't remove all the adapters
when, for example, one of the reads is of poor quality/has a lot of N base calls,
so that trimmomatic doesn't manage to identify the relevant parts of R1 and R2
as reverse complements of each other.

What does the trimmomatic output say about how many reads are being trimmed by the palindrome clipping?

Are the plots you posted from R1 or R2?

[jasbhead]

Are you only doing palindrome clipping?

How long are the adapter sequences for palindrome clipping in your IlluminaClipping.fasta file?

I find that sometimes palindrome clipping alone doesn't remove all the adapters
when, for example, one of the reads is of poor quality/has a lot of N base calls,
so that trimmomatic doesn't manage to identify the relevant parts of R1 and R2
as reverse complements of each other.

What does the trimmomatic output say about how many reads are being trimmed by the palindrome clipping?

Are the plots you posted from R1 or R2?

Turns out I was not doing palindrome initially. I am doing palindrome cutting now and it appears to be cutting more of the adapter sequences, leaving only a few bp on the 3' end. Attached are the new plots for R1 and R2. In what I assume to be a separate issue, R2 reads seem to have enrichment for CCCCC and GGGGG -mers. No idea what those could be from. Advice on that would also be appreciated.

IlluminaClipping.fa:
>PrefixPE/1
ACACTCTTTCCCTACACGACGCTCTTCCGATCT
>PrefixPE/2
AGATCGGAAGAGCACACGTCT

Output from IlluminaClip:
Using PrefixPair: 'ACACTCTTTCCCTACACGACGCTCTTCCGATCT' and 'AGATCGGAAGAGCACACGTCT'
ILLUMINACLIP: Using 1 prefix pairs, 0 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences
Input Read Pairs: 4000000 Both Surviving: 2364231 (59.11%) Forward Only Surviving: 1635769 (40.89%) Reverse Only Surviving: 0 (0.00%) Dropped: 0 (0.00%)

[mastal]

Have a look at the most recent version of the trimmomatic manual,
there is a min adapter length parameter that you can set for the Illuminaclip command, that will remove shorter adapter sequences.

I just remembered another reason why trimmomatic sometimes doesn't
get all the adapter sequences - when there are more than 2 (or whatever the specified value is) mismatches in the seed region.

As for the stretches of G's and C's, I'm not sure, they might be artifacts from the flow cell, sometimes you see sequences like that with reads where the insert is very short and you read all the way through the end of the adapter sequences.

[GenoMax]

As for the stretches of G's and C's, I'm not sure, they might be artifacts from the flow cell, sometimes you see sequences like that with reads where the insert is very short and you read all the way through the end of the adapter sequences.

Not sure if that explanation is applicable in this case as they are present only in R2.

@jabshead: What kind of an experiment is this? Are you expecting enrichment of GC rich sequences at the R2 end of the fragments? What is the insert size?

[jasbhead]

Have a look at the most recent version of the trimmomatic manual,
there is a min adapter length parameter that you can set for the Illuminaclip command, that will remove shorter adapter sequences.

I just remembered another reason why trimmomatic sometimes doesn't
get all the adapter sequences - when there are more than 2 (or whatever the specified value is) mismatches in the seed region.

As for the stretches of G's and C's, I'm not sure, they might be artifacts from the flow cell, sometimes you see sequences like that with reads where the insert is very short and you read all the way through the end of the adapter sequences.

The min adapter length was the issue! Had to set it to 1 (default was 10 or something). thanks for the advice. FastQC for the R1 reads show no k-mer enrichment but the G's and C's issue persists in the R2 reads, as shown in the attached picture.

Not sure if that explanation is applicable in this case as they are present only in R2.

@jabshead: What kind of an experiment is this? Are you expecting enrichment of GC rich sequences at the R2 end of the fragments? What is the insert size?

This is a WG resequencing experiment in maize. No, I was not expecting GC rich sequences in the R2 reads, unless I am missing something. The median fragment size was 270bp (mistakenly chosen by the genomics facility, I presume), so I was expecting a substantial amount of read-through since the reads are 150bp.