Computational tools for predicting novel substrates for a class of enzymes

Question

[yarrowmadrona]

Does anyone have experience with computational tools that would allow prediction of novel substrates for a given class of enzymes? For example, something that good give confidence in predicting novel combinations of enzymes/substrates. A requirement is that the substrate is turned over by the enzyme with the same chemical modifications occurring in the novel substrate as occurs in the known substrate.

I have found: https://github.com/AlexanderKroll/ESP https://esp.cs.hhu.de/

But I have not used this tool. Thank you.

Answer 1

[hakimbazol]

have you try DeepEC before? actually this tool is not directly show you the novel substrate, but based on protein sequence that you input, this tool can make a prediction about EC number and based on this, the novel substrate might be discovered.

[yarrowmadrona]

hakimbazol Thankyou for our input. Sorry about the late reply. Was off-line a bit with the Thanksgiving holiday. Looks promising. Have you tried ESP? https://www.nature.com/articles/s41467-023-38347-2 I have not but looking for people who have used it.

[hakimbazol]

It is just a new tool for me, so I did not use it yet. However, DeepEC has been used several times I think, so I think it also means the robustness of the DeepEC.