FAQs
No, you can enter only sequences of length 1024. Butyou may truncate the input sequence to reduce its length(for e.g. truncate the N-terminal such that the length of the truncated sequence is <= 1024.
No, this feature is not yet included in the model.
No, there is no in-built feature PRECOGx to perform any in-silico design. However, PRECOGx can predict the coupling probabilities of any GPCR sequence. So the user can perform an in-silico design by themselves and then feed the mutated sequence to PRECOGx as input.
Yes, please check out our GitHub page here.
No, PRECOGx only accepts point mutations in the mutation format style. However, the user can generate sequences (FASTA-formatted) with more than one insertion (or deletions), and feed them to PRECOGx as input.
No, PRECOGx was trained and tested only by considering only GPCR sequences. PRECOGx will eliminate any non-GPCR sequence(s) from the input.
Yes, PRECOGx can make predictions for all the classes of GPCRs. However, the user must remember that the majority of receptors used to develop PRECOGx belong to Class A.
PRECOGx displays only the 3D structure complexes of GPCR/G-proteins or GPCR/Beta-arrestins available in PDB. Currently, PRECOGx does not display 3D structures with only GPCRs as polypeptides.
We update PRECOGx structures once every month. If you are looking for a structure that was released over a month ago, please drop us an email, and we will make it available as soon as we can.
Currently, PRECOGx displays only the 3D structure complexes with a GPCR/G-protein pair that is known to be coupled in TGF/GEMTA experimental assays. We didn’t predict GPCR/β-arrestins complexes as one of the key factors determining β-arrestin recruitment, i.e. PTM (post translation modification), cannot yet be included in AlphaFold2 predictions. In the future, we plan to introduce AlphaFold-predicted complexes for any input sequence.
No, these numbering schemes are different. Check this page to know more.
Currently, PRECOGx displays only the GPCRdb numbering scheme. In the future, we plan to introduce other schemes too.
No, this feature is not yet available in PRECOGx. But we understand the importance of this feature and plan to introduce it in the next version.
We find the best 3D structure for input by computing the sequence identity (using PSI-BLAST) of the input against all the 3D structure complexes available from PDB and predicted by AlphaFold. The 3D structures are ranked in decreasing order of sequence identity, and the one with the highest value is displayed by default. As a result, you may still see a 3D structure for your input even from a homologous template.