search

sokrypton / ColabFold

Making Protein folding accessible to all!
MIT License
1.87k stars 473 forks source link

How to dock peptide to protein with ColabFold #64

Open gundalav opened 2 years ago

gundalav commented 2 years ago

Hi,

Suppose I have a Peptide IQGLSFDIYEALLwhich I want to dock (create a heterocomplex?) to a protein QEEACAKNFGAFCVFVGIVRKEDN

Is this syntax correct to run with AlphaFold2_advanced ?

QEEACAKNFGAFCVFVGIVRKEDN:IQGLSFDIYEALL Notice the colon (:)

Especially in the sequence entry. For homooligomer entry I simply leave it as it is. Thanks and hope to hear from you again.

G.V.

milot-mirdita commented 2 years ago

It doesn't seem to return any errors for me. The prediction is just not very good. You can try tweaking the advanced options to see if it eventually produces better structures. The two sequences might be just too short though, especially the second sequence is running into MMseqs2 issues with minimum sequence length.

gundalav commented 2 years ago

Hi Milot, Thank you for replying. That was just an example. In reality I use this as input to dock short peptide on 500aa receptor.

CSNLCLLSPGGGHKCACPTNF …. (another 478 aa, Lrp1.Af2.m1) : MGKGDPKKPRGKMSSYAFFVQTCREEHKKKHPDASVNFSEFSKK

I don't see any contact between the two:

download

What parameter you suggest I can play with to improve the result?

G.V.

milot-mirdita commented 2 years ago

You can try out the various pairing modes or increase the number of recycle iterations (max_recycles). I think these are the two most important parameters. Could you give us a bit more details what these sequences are? You could also write an email if you cannot share that publicly.

gezmi commented 2 years ago

Hi! We tested various parameters for peptide-protein docking. It is advisable to try both with a 30-residue-long polyG-linker and as separate chains, as you did. We got best performance with setting recycles higher (9). Still, in several cases, the peptide simply does not touch the receptor. Maybe AF-multimer will help those.

gundalav commented 2 years ago

Hi! We tested various parameters for peptide-protein docking. It is advisable to try both with a 30-residue-long polyG-linker and as separate chains, as you did. We got best performance with setting recycles higher (9). Still, in several cases, the peptide simply does not touch the receptor. Maybe AF-multimer will help those.

Hi Julia,

Thank you so much for your reply. How can I add and format my fasta with a 30-residue-long polyG-linker ? Let say I have these which I'd like to dock with AF2.

e.g. receptor: ACPTNFYLGSDGRTCVSNCTASQFVCKNDKCIP ligand : CSNLCLLSPGGGHKCDGTDNCGDGTD

Currently with hetero-complex mode I format the input the following way (notice the colon)

> my input
ACPTNFYLGSDGRTCVSNCTASQFVCKNDKCIP:CSNLCLLSPGGGHKCDGTDNCGDGTD

Is this how I should add poly-G linker:

> my_input_with_polyG
ACPTNFYLGSDGRTCVSNCTASQFVCKNDKCIPGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG:CSNLCLLSPGGGHKCDGTDNCGDGTD

Finally where can I find the AF2-Multimer under ColabFold?

Thanks and hope to hear from you again.

G.V.