G. Uguzzoni

Unsupervised modeling of mutational landscapes of adeno-associated viruses viability

This paper introduces a biophysically-inspired model to predict the viability of AAV2 capsid variants using DMS data, evaluating its performance on log-selectivity and binary classification.

Inference and design of antibody specificity: from experiments to models and back

This paper develops a multi-stage approach combining phage display experiments, machine learning, and biophysical modeling to design antibodies with customized specificity profiles.

Inference of annealed protein fitness landscapes with AnnealDCA

This paper introduces AnnealDCA, a machine-learning framework to infer protein fitness landscapes from sequencing data, demonstrating its effectiveness on antibody Rep-Seq and screening experiments.

AMaLa: analysis of Directed Evolution experiments via Annealed Mutational approximated Landscape

AMaLa infers fitness landscapes from Directed Evolution experiments by modeling mutation and selection dynamics across sequenced rounds, outperforming existing methods.