Adapter existence only in _2 RNA-seq data

[netpumber]

Hi.

While i was looking at some FASTQC results, i realized that in test_1.fq file FASTQC didn't find any adapter or over-expressed sequence. Although on the other hand FASTQC found illumina's adapter on test_2.fq file.

1) I thought that illumina's machine at the end of the process, removes automatically all adapter sequences while it seems that it doesn't

2) Also i realized that there are reads constituting by sequences like this:

adapterSequence--adapterSequence--readDequence

Is that possible ? Is it possible two or more adapter sequences merged together and then read start ? How that bias occurred ?

3) Finally is there any way of getting out the 100% of the reads ? Nothing less and nothing more. Just the reads. I'm asking because it finally seems that inside the "reads" there are sequences that have nothing to do with real reads.

Any tool/article/video will be helpful.

Thank you.

[GenoMax]

Hi.

1) I thought that illumina's machine at the end of the process, removes automatically all adapter sequences while it seems that it doesn't

Only on MiSeq's, on-board analysis software (if it is being used), has the option of removing adapters (if they are present in the read) as a part of pre-processing.

3) Finally is there any way of getting out the 100% of the reads ? Nothing less and nothing more. Just the reads. I'm asking because it finally seems that inside the "reads" there are sequences that have nothing to do with real reads.

That is the reason trimming programs exist. Popular options are BBDuk (from BBMap), trimmomatic, cutadapt (and others). All have threads of their own on SeqAnswers.