Paesani, F. | Search bioRxiv

bioRxiv2025

Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition

In this report, we summarize and analyze the 2024 Adaptyv protein design competition. Participants used computational and Machine Learning (ML) methods of their choice to design proteins that bind the Epidermal Growth Factor Receptor (EGFR), a key drug target involved in cell growth, differentiation, and cancer development. Over 1,800 designs were submitted across two rounds. Of these, 601 proteins were selected and characterized for expression and binding affinity to EGFR, with competitors both optimizing existing binders (KD = 1.21 nM) and creating de novo binders (KD = 82 nM). All selected designs were experimentally validated using Adaptyvs automated Bio-Layer Interferometry (BLI) pipeline. This competition illustrates the potential of crowdsourcing to drive creativity and innovation in protein design. However, it also exposed key challenges, such as the lack of standardized benchmarks, experimental design targets, and robust computational metrics for method comparison. We anticipate that future competitions will address these gaps and further motivate progress in computational protein design.

bioengineering↗

bioRxiv2025

Computational Analysis of Threonine Ladders on Distinct Beta-Solenoid Scaffolds, with Implications for the Design of Novel Antifreeze Proteins

Cold-adapted organisms frequently express antifreeze proteins (AFPs) that facilitate their survival at low temperatures, with some especially potent insect AFPs exhibiting beta-solenoid structures with ice-binding threonine ladders. Beta-solenoids exist in nature in numerous forms and emerging protein design technologies may afford opportunities to diversify them further, suggesting the possibility of developing a variety of new AFPs by installing a threonine ladder on non-AFP natural or designed beta-solenoids. However, early attempts at such engineering, combined with differences observed between AFPs and structurally similar ice-nucleating proteins, have raised a critical question: Will a threonine ladder show essentially the same behavior regardless of the beta-solenoid scaffold that hosts it, or does the specific solenoid scaffold significantly affect a threonine ladders structural characteristics (and thus potentially alter its suitability for ice binding)? We set out to address this question by creating distinct variants of a simplified model beta-solenoid for in silico analysis via structure prediction and molecular dynamics simulations. Our findings indicate that local structural details such as the distance between the hydroxyl groups of adjacent threonines in a TXT motif can vary depending on the beta-solenoid scaffold. In the most extreme example among our model solenoids, we observed in simulations that differences in only inward-facing residues of the scaffold were sufficient to influence the presence of ordered channel waters between the threonines, a noted feature of natural ice-binding threonine surfaces such as that of TmAFP. While additional studies will be necessary to expand on how such distinctions affect activity, these results emphasize that the impact of a particular beta-solenoid scaffold on the local geometry of a threonine ladder may be a pertinent consideration in future efforts to design novel hyperactive AFPs to support applications ranging from biomedical cryopreservation to food science. We conclude our present investigation with a preliminary exploration of how this and other considerations manifest in a proposed workflow for generating predicted AFP-like beta-solenoids using AlphaFold and ProteinMPNN.

biophysics↗