over-represented is not removed after trimmomatic

Question

[ATCG]

I am trimming over-represented sequences using the following:

#make .fa with sequence found to be over-represented
echo ">illumina" > adapter.fa
echo "AGATCGGAAGAGCACACGTCAGAACTCCAGTCACAACCAGATCTCGTATG" >> adapter.fa

#Run trimmomatic using the adapter.fa
trimmomatic SE WT_1.fastq.gz WT_1.trim.fastq.gz ILLUMINACLIP:adapter.fa:2:30:5

#Run Fastqc
fastqc WT_1.trim.fastq.gz

After running fastqc on the trimmed file I still have the same over-represented sequence

TrimmomaticSE: Started with arguments:
 WT_1.fastq.gz WT_1.trim.fastq.gz ILLUMINACLIP:adapter.fa:2:30:5
Automatically using 1 threads
ILLUMINACLIP: Using 0 prefix pairs, 0 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences
Quality encoding detected as phred33
Input Reads: 21296189 Surviving: 21296189 (100.00%) Dropped: 0 (0.00%)
TrimmomaticSE: Completed successfully
(bioinfo)

[lieven.sterck]

could you perhaps share (part of) the fastQC plots which hinted you at this over representation?

and not sure but I think that trimmomatic (as well many other software) will only clip if the 'sequence' is at the extremities of the read (?)

[ATCG]

The problem that my adapter.fa file was not in the path: ILLUMINACLIP: Using 0 prefix pairs, 0 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences

when it works TrimmomaticSE: Started with arguments: WT_1.fastq WT_1.trim.fastq ILLUMINACLIP:adapter.fa:2:30:5 Automatically using 1 threads

Using Long Clipping Sequence: 'AGATCGGAAGAGCACACGTCAGAACTCCAGTCACAACCAGATCTCGTATG' ILLUMINACLIP: Using 0 prefix pairs, 1 forward/reverse sequences, 0 forward only sequences, 0 reverse only sequences Quality encoding detected as phred33