Ajasja Ljubetič

The paper validates and analyzes 12,000 AI-designed CAR-T binders from a global competition, identifying successful designs and factors improving their performance.

Prosculpt is a protein design pipeline that simplifies the use of multiple tools for protein design, enabling users to perform complex tasks without coding.

The paper designs and characterizes a random protein walker that diffuses along a designed protein track, finding that walkers with more feet move faster.

The paper introduces a method to control protein activity through dimerization-based conditional reconstitution, demonstrating its application in inducing cell death, virus detection, and immune cell stimulation.

The paper introduces a system to regulate protein function and create Boolean logic gates using coiled-coil dimers in protein loops, demonstrated on ten diverse proteins.

The paper develops a high-throughput method to characterize protein-protein interactions and design orthogonal coiled-coil pairs, identifying a large set of specific interactions.

The paper introduces MP3-seq, a scalable method for measuring protein-protein interactions, and shows it can quantify over 100,000 interactions.

The paper designs and characterizes trivalent miniprotein binders that neutralize SARS-CoV-2 variants with low pM IC50 values and provide prophylactic and therapeutic protection in mice.

This paper introduces Scrambler networks, a method for interpreting sequence-based neural networks by learning stochastic masks to identify salient sequence positions and explain model predictions.

The paper presents a framework for ensuring reproducibility in bio-macromolecular modeling through automated benchmarks and a test server, demonstrating its utility for Rosetta and other tools.