M. Cagiada

Predicting absolute protein folding stability using generative models

The paper uses generative models to predict absolute protein stability, achieving a mean error of 1.5 kcal/mol and a correlation coefficient of 0.7.

A mutational atlas for Parkin proteostasis

This study maps Parkin missense variants' abundance and degradation mechanisms, showing most low-abundance variants are degraded via the ubiquitin-proteasome system and stabilized at lower temperatures.

Characterizing glucokinase variant mechanisms using a multiplexed abundance assay

The paper characterizes mechanisms of human glucokinase variants by analyzing their abundance and stability, identifying residues that affect metabolic stability and conformational dynamics.

Rare catechol-O-methyltransferase (COMT) missense variants are structurally unstable proteasome targets

The paper shows two rare COMT missense variants are structurally unstable, leading to proteasomal degradation and reduced enzyme levels, with degradation independent of COMT isoform.

Lysine deserts prevent adventitious ubiquitylation of ubiquitin-proteasome components

The paper identifies lysine deserts in UPS proteins and shows that introducing lysines increases ubiquitylation, with functional consequences for protein stability and function.

Effect of naturally-occurring mutations on the stability and function of cancer-associated NQO1: comparison of experiments and computation.

The paper compares computational predictions and experimental analysis of NQO1 variants to assess their stability and function, finding good correlation but noting outliers.

Rapid protein stability prediction using deep learning representations

This paper introduces RaSP, a deep learning method for rapid protein stability prediction, showing it matches biophysics-based methods and reveals population-level patterns in stability changes.

Discovering functionally important sites in proteins

This paper introduces a machine learning method to predict functional sites in proteins by integrating sequence conservation and biophysical stability models, validated through experimental data on missense variants.

Predicting and interpreting large scale mutagenesis data using analyses of protein stability and conservation

The paper predicts and interprets large-scale mutagenesis data by analyzing protein stability and conservation, finding that sequence conservation provides more accurate variant effect predictions than stability calculations.

Understanding the origins of loss of protein function by analyzing the effects of thousands of variants on activity and abundance

The paper analyzes variant effects on protein activity and abundance in NUDT15 and PTEN, finding that a third of variants cause loss of function and half of these also reduce abundance.