I have full proteome data (label-free MS/MS) from solid tissue samples of breast cancer and i heve already identified the canonical proteins expressed. Now, I am trying to identify peptides originated by "non-coding" RNAs...

The problem is, from 1519 ncRNA-coded peptides identified, only 44 are not aligned to canonical protein sequences.

Is there any way to determine if the peptide is really only a "fragment" derived from a canonical protein, or if it is a peptide from a ncRNA? Maybe a sequence coverage weighting, or any abundance comparison?

(P.S - The list of candidates ncRNA-coded peptides come from predicted and validated studies)