Biology subjects

Izadi, S.

Publications and source records attributed to Izadi, S..

6 recordsLinked to original sources

bioRxiv2026

Rapid generation of antibody conformational ensembles using machine learning

We introduce Cadense, a generative framework for antibody ensemble prediction that produces diverse Fv structures from sequence, informed by conformational dynamics. While a valuable resource, static structure prediction provides only a single snapshot of dynamic proteins, whose motions influence their binding and other properties. This is particularly relevant for antibodies, an important class of molecules for drug discovery, where flexible complementarity determining region (CDR) loops are difficult to model with static approaches. Cadense addresses this gap by generating ensembles that match conformational and biophysical property distributions from molecular dynamics (MD), while reducing the computational burden from days of simulation to seconds of generation. We show Cadense ensembles can recover both apo and holo conformations of the same antibody and serve as effective input to physics-based docking, capturing the conformational flexibility important to properly model antibody antigen interactions. Together, these results establish Cadense as a useful tool for antibody conformational ensemble generation with the promise to enable faster, dynamics-aware antibody design.

bioinformatics↗

bioRxiv2026

Self-Interaction Nanoparticle Spectroscopy Predicts High-Concentration Viscosity of Therapeutic IgG1 Antibodies

Predicting high-concentration viscosity of monoclonal antibodies such as IgG1 is crucial for their development as therapeutics for subcutaneous delivery. Unfortunately, traditional experimental rheometry methods for assessing viscosity are low-throughput. This study evaluates Self-Interaction Nanoparticle Spectroscopy (SINS) assays--specifically charge-stabilized SINS (CS-SINS) and PEG-stabilized SINS (PS-SINS)--for high-throughput viscosity prediction. We characterized 96 IgG1 antibodies, assessing SINS against in silico descriptors and dynamic light scattering (DLS) data. CS-SINS showed strong correlation with charge, offering limited additional utility. In contrast, PS-SINS provided orthogonal information; integrating it with in silico data and DLS significantly improved random forest model accuracy for binary viscosity classification. PS-SINS measurements in multiple buffers captured complementary information, achieving comparable accuracy without DLS. Importantly, PS-SINS scores exhibited a strong logarithmic relationship (r=0.98) with high-concentration viscosity in Fc variants of clinical antibodies, suggesting a direct mechanistic link. Furthermore, PS-SINS performed reliably with one column purified (protein A) samples, supporting its early-stage application. These findings establish PS-SINS as a high-throughput tool to accelerate the developability assessment of antibody candidates.

biochemistry↗

bioRxiv2025

Rational design of oxidation-resistant antibodies through local electrostatic modulation

Oxidation is a significant degradation pathway in proteins, particularly therapeutic antibodies, that can impair function, efficacy, and stability. Understanding the sequence and structural factors that drive oxidation susceptibility is critical for incorporating chemical stability into early-stage antibody design. Here, we present a machine learning classifier trained on expert-guided structural features to assess tryptophan (Trp) oxidation risks in the complementarity-determining regions (CDR) of 187 antibodies produced internally at Genentech. The model reveals a strong correlation between local negative electrostatic potential and Trp oxidation susceptibility. A simplified two-parameter model derived from these insights achieves 79% accuracy in classifying oxidation risk, compared to 84% accuracy with the full-feature classifier. Beyond our internal dataset, the two-parameter model successfully predicted oxidation risk for all CDR Trp sites in a blind subset of eight clinical-stage antibodies. In addition, we show that modulating the electrostatic potential around the Trp side-chains through distant charge-altering mutations can significantly reduce oxidation rates. In four out of five re-engineered antibodies, oxidation rates decreased by approximately 50%, while half of these maintained binding affinity. Finally, we demonstrate that this approach can guide multi-property optimization by balancing oxidation resistance and affinity in an anti-CD33 antibody. These results establish a strong link between local electrostatic environments and Trp oxidation susceptibility, and provide a practical framework for designing oxidation resistant biotherapeutics.

biochemistry↗

bioRxiv2024

Leveraging protein language and structural modelsfor early prediction of antibodies with fast clearance

Monoclonal antibodies (mAbs) with long systemic persistence are widely used as therapeutics. However, antibodies with atypically fast clearance require more dosing, limiting their clinical usefulness. Deep learning can facilitate using sequence-based modeling to predict potential pharmacokinetic (PK) liabilities before antibody generation. Assembling a dataset of 103 mAbs with measured nonspecific clearance in cynomolgus monkeys (cyno), and using transfer learning from large protein language models, we developed multiple machine learning models to predict mAb clearance as fast/slow clearing. Focusing on minimizing misclassification of potentially promising molecules as fast clearing, our results show that using physicochemical properties yielded up to 73.1+/-1.1% classification accuracy on hold-out test data (precision 65.2+/-2.3%). Using only sequence-based features from deep learning protein language models yielded a comparable performance of 71+/-1.4% (precision 65.5+/-2.5%). Combining structural and deep learning derived features yielded a similar accuracy of 73.9+/-1.1%, and slightly improved precision (68.3+/-2.4%). Features important for classifying fast/slow clearance point to charge, moment, and surface area properties at pH 7.4 as well as deep learning derived features. These results suggest that the protein language models provide comparable information and predictive performance of clearance as physicochemical features. This work provides a foundation for in silico prediction of protein pharmacokinetics to inform antibody candidate generation and early deprioritization of designs with high risk of fast clearance. More generally, it illustrates the value of transfer learning-based application of protein language models to address characteristics of importance for protein therapeutics.

pharmacology and toxicology↗

bioRxiv2023

Molecular Surface Descriptors to Predict Antibody Developability

Understanding the molecular surface properties of monoclonal antibodies (mAbs) is crucial for determining their function, affinity, and developability. Yet, robust methods to accurately represent the key structural and biophysical features of mAbs on their molecular surface are still limited. Here, we introduce MolDesk, a set of molecular surface descriptors specifically designed for predicting antibody developability characteristics. We assess the performance of these descriptors by directly benchmarking their correlations with an extensive array of in vitro and in vivo data, including viscosity at high concentration, aggregation, hydrophobic interaction chromatography (HIC), human pharmacokinetic (PK) clearance, Heparin retention time, and polyspecificity. Additionally, we investigate the sensitivity of these surface descriptors to methodological nuances, such as the choice of interior dielectric constant for electrostatic potential calculations, residue-level hydrophobicity scales, initial antibody structure models, and the impact of conformational sampling. Based on our benchmarking analysis, we propose six in silico developability rules that leverage these molecular surface descriptors and demonstrate their superior ability to predict the clinical progression of therapeutic antibodies compared to established models like TAP. 1

bioinformatics↗

bioRxiv2021

Divining Deamidation and Isomerization in Therapeutic Proteins: Effect of Neighboring Residue

Deamidation of asparagine (ASN) and isomerization of aspartic acid (ASP) residues are among the most commonly observed spontaneous post-translational modifications (PTMs) in proteins. Understanding and predicting a protein sequences propensity for such PTMs can help expedite protein therapeutic discovery and development. In this study, we utilized proton-affinity calculations with semi-empirical quantum mechanics (QM) and {micro}s long equilibrium molecular dynamics (MD) simulations to investigate mechanistic roles of structure and chemical environment in dictating spontaneous degradation of asparagine and aspartic acid residues in 131 clinical-stage therapeutic antibodies. Backbone secondary structure, side-chain rotamer conformation and solvent accessibility were found as three key molecular indicators of ASP isomerization and ASN deamidation. Comparative analysis of backbone dihedral angles along with N-H proton affinity calculations provides a mechanistic explanation for the strong influence of the identity of the n+1 residue on the rate of ASP/ASN degradation. With these findings, we propose a minimalistic physics-based classification model that can be leveraged to predict deamidation and isomerization propensity of therapeutic proteins.

biochemistry↗