there is a bit too little detail in your post to tell I'm afraid.

But do take into account that fastQC is very stringent (and it's base line is DNAseq), so that you will only have green checks is rather the exception than rule, especially when inputting RNAseq data. (eg. for RNAseq, due to its nature you will always have the red check for the duplication levels section)

Hi,

Without showing the plots it is difficult to comment. I would say that (depending also on the library prep) it is normal to have a warning/error (orange/read) flags on Per base sequence content, specially a bias in the beginning that depends on the library prep, as well as Sequence Duplication Levels due to some intrinsic high transcripts and/or PCR amplification.

I would say that the other warnings/error flags can be more worrisome, but of course it depends on the library and the data. For instance the flag Overrepresented sequences can be due to some adapters that you did not remove, etc.

Read more about these details and interpretations in this nice tutorial/course: hbctraining

I hope this helps,

António