Bian Li

PROPERMAB predicts antibody developability properties using machine learning, enabling early assessment of biophysical characteristics to accelerate drug development.

The paper identifies a novel SCN5A variant, E171Q, that creates a gating pore current by neutralizing a negative charge in the S2 helix, leading to MEPPC and altered voltage dependence.

This study maps the functional effects of thousands of KCNE1 variants using high-throughput methods, identifying loss-of-function and loss-of-surface expression variants and showing concordance with electrophysiological and computational data.

The paper introduces the EN-TEx resource, which provides personal epigenomes and variant-impact models across multiple tissues, revealing allele-specific loci and their associations with GWAS loci.

The paper uses personalized structural biology to analyze de novo KCNC2 variants in a DEE patient, revealing distinct functional effects of two nearby variants.

The paper shows that most SCN5A missense loss-of-function variants cause dominant negative effects and are associated with higher Brugada Syndrome risk.

The paper introduces COSMIS, a framework to quantify 3D spatial constraint on amino acid sites in the human proteome, showing it outperforms existing metrics and links constraint to disease relevance.

The paper introduces MASSP, a deep-learning system that simultaneously predicts secondary structure, transmembrane segments, and topology of proteins, outperforming existing methods in accuracy and applicability.

This paper introduces ThermoNet, a 3D-convolutional neural network that predicts mutation-induced changes in protein thermodynamic stability, showing balanced performance on stabilizing and destabilizing mutations.

This study reclassifies SCN5A variants using high-throughput patch clamping, finding most variants of uncertain significance are either benign or pathogenic.