Conserving the end of a protein sequence
Hello!
I was wondering if it is possible to generate a sequence by conserving the end of the sequence (rather than the beginning, which is possible if you provide the beginning of the sequence for ProtGPT to complete).
I have seen some other protein generation models which were trained using each the sequence and it's reverse order - if this is the case then I think I could just provide the end of the sequence in reverse order to generate new proteins - but I'm not sure if this is how ProtGPT was trained.
Any other thoughts on how one would go about conserving the end of a sequence?
Thanks!
Kathryn
Hi Kathryin,
No, it is not possible I'm afraid. ProtGPT2 was only trained in the direction N->C terminus.
Have a nice day
Noelia
Got it, thank you.
Kathryn