Protein language model identifies disordered, conserved motifs implicated in phase separation.

Area of Science:

• Biochemistry
• Molecular Biology
• Evolutionary Biology

Background:

• Intrinsically disordered regions (IDRs) are vital for protein phase separation and the formation of membraneless organelles (MLOs).
• Mutations in IDRs can disrupt molecular interactions, leading to altered phase behavior and disease.
• Understanding evolutionary constraints on IDRs offers insights into sequence-based phase separation.

Purpose of the Study:

• To investigate the evolutionary conservation of amino acids within IDRs.
• To map residue-level mutational tolerance landscapes in IDRs using protein language models.
• To correlate evolutionary constraints with the role of IDRs in phase separation and MLO formation.

Main Methods:

• Utilized the ESM2 protein language model to predict residue-level mutational tolerance.
• Analyzed multiple sequence alignments to confirm evolutionary constraints.
• Examined conserved amino acids, including 'stickers' and 'spacers', within IDRs.

Main Results:

• IDRs involved in phase separation exhibit significant amino acid conservation.
• ESM2 predictions of mutational constraints align with direct sequence analyses.
• Conserved amino acids often form continuous sequence motifs, acting as both 'stickers' and 'spacers'.

Conclusions:

• Conserved motifs within IDRs are likely functional units under evolutionary selection for stable MLO formation.
• Evolutionary analysis, aided by protein language models, reveals the molecular grammar of phase separation.
• This study highlights the interplay between protein sequence, evolution, and the biophysical properties of IDRs.