A simple workflow to identify novel small linear motif (SLiM)-mediated interactions with AlphaFold.

Area of Science:

• Computational biology and bioinformatics
• Molecular biology and biochemistry

Background:

• Short linear motifs (SLiMs) are crucial for cellular homeostasis but are challenging to detect computationally and experimentally.
• Existing methods for SLiM detection often rely on structure-derived benchmarks, limiting their applicability.

Purpose of the Study:

• To evaluate AlphaFold (AF), specifically AF2 and AF3 via ColabFold, as a tool for in-silico screening of SLiMs.
• To develop and validate a novel, accessible workflow for SLiM identification and experimental validation.

Main Methods:

• Assessed AF2 and AF3 using a structure-independent benchmark of 26 interactions.
• Utilized MiniPAE as the most suitable AlphaFold metric for SLiM screening.
• Developed a SLiM screening strategy with an adaptive scoring threshold to address AlphaFold's specificity limitations.
• Introduced a proximity labeling method for experimental SLiM validation in living cells.

Main Results:

• MiniPAE demonstrated suitability for SLiM screening with AlphaFold.
• An unbalanced dataset revealed limitations in AlphaFold's specificity for SLiM detection.
• The developed workflow successfully identified and experimentally validated 13 new SLiMs binding to RPS6KA3 (RSK2).

Conclusions:

• AlphaFold, particularly with ColabFold and MiniPAE, provides a practical and accessible method for in-silico SLiM screening.
• The proposed strategy and validation method overcome key challenges in SLiM discovery.
• This approach offers a scalable and widely accessible solution for identifying functional SLiMs in proteins of interest.