removing adaptor

I have two RNAseq datasets, from two different sequencing core facilities – the fastq files they provide already are demultiplexed and trimmed for adaptor.

This is the Adaptor content following fastqc-> multiqc . According to the multiqc all files passed the qc for adapter content. So I was under impression that I dont have to remove adaptor from them.

Then I got the second dataset from a different facility. And their dataset does not show even a graph suggesting the adaptor content of each files are < 0.1%.

This makes me think, should I redo all my analysis with removing adaptors to the level of <0.1% in my first dataset too ?

In strict sense you don’t need to trim the adapter if you are aligning to a good reference genome. Aligners will soft-clip parts of the read that do not align. So you should be fine either way. If you are planning to do any de novo assembly work then you should remove all extraneous sequence, which will include the adapter sequences.