LazyAF, a pipeline for accessible medium-scale in silico prediction of protein-protein interactions
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Arcadia Science)
- Articles for Josy (josyphiri)
Abstract
Artificial intelligence has revolutionized the field of protein structure prediction. However, with more powerful and complex software being developed, it is accessibility and ease of use rather than capability that is quickly becoming a limiting factor to end users. Here, I present a Google Colaboratory-based pipeline, named LazyAF, which integrates the existing ColabFold BATCH to streamline the process of medium-scale protein-protein interaction prediction. I apply LazyAF to predict the interactome of the 76 proteins encoded on a broad-host-range multi-drug resistance plasmid RK2, demonstrating the ease and accessibility the pipeline provides.
Availability
LazyAF is freely available at https://github.com/ThomasCMcLean/LazyAF
Article activity feed
-
I really appreciate your answers to my questions, thanks for taking the time to respond!
-
No worries, I assumed but just wanted to make sure!
-
Makes sense, thank you!
-
Good question, i need to double check but im 99% certain it is chain-location independent.
-
I mention it briefly in the text but adding it to the figure is a great idea, thank you!
-
Sorry, it was there in the original submission pdf but bioRxiv seems to have cut it off, i'll get that fixed. Orange is >0.8 and deep purple is 0-0.2
-
Absolutely. If you have a few in mind for one complex i would just put that all into the original ColabFold as a single job, otherwise I would use the results from an in silico pulldown to guage what might be forming a complex using a networking tool such as cytoscape and then go from there
-
It can but currently this requires doing each bait seperately. e.g. protein A vs. list A then protein B vs. list A and on. The main limiting factor is the computing time and cost in Part 2
-
Thanks, i hope you find it easy to use and useful!
-
Here, I present a Google Colaboratory-based pipeline, named LazyAF, which integrates the existing ColabFold BATCH to streamline the process of medium-scale protein-protein interaction prediction.
Thanks so much for sharing this tool! I really enjoyed reading about it and can't wait to try it myself.
-
Supplementary Note 1
I really love this walkthrough!!
-
Figure 3.
You mentioned that several of the top predicted PPIs have been previously experimentally validated. It would be interesting to know which of these annotations predicted by your pipeline are supported by experimental validation. Could you somehow denote this in this figure?
-
Figure
Could you include a key with the color scale for the heatmap?
-
uggesting that a score of co-folding between protein A (bait):protein B (candidate) sometimes is different from that of a co-folding between protein B (bait):protein A (candidate).
If you run the analysis multiple times, do you see variation between runs or is it only when you switch the chains?
-
the sequence of the ‘bait’ and that of a ‘candidate’ joined via a colon.
I'm curious if you're pipeline supports more than 2 proteins. For example, could I put in a handful of proteins to see if they form a complex?
-
a ‘bait’ protein
I think based on your analysis that the answer to this question is yes, but can you do multiple bait proteins as well as multiple candidate proteins? Also curious about how many proteins this can handle?
-