Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AbAffinity: A Large Language Model for Predicting Antibody Binding Affinity against SARS-CoV-2

Faisal Bin Ashraf, Animesh Ray, Stefano Lonardi

TL;DR

Ab-Affinity is introduced, a new large language model that can accurately predict the binding affinity of antibodies against a target peptide, e.g., the SARS-CoV-2 spike protein.

Abstract

Machine learning-based antibody design is emerging as one of the most promising approaches to combat infectious diseases, due to significant advancements in the field of artificial intelligence and an exponential surge in experimental antibody data (in particular related to COVID-19). The ability of an antibody to bind to an antigens (called binding affinity) is one of the the most critical properties in designing neutralizing antibodies. In this study we introduce Ab-Affinity, a new large language model that can accurately predict the binding affinity of antibodies against a target peptide, e.g., the SARS-CoV-2 spike protein. Code and model are available at https://github.com/ucrbioinfo/AbAffinity.

AbAffinity: A Large Language Model for Predicting Antibody Binding Affinity against SARS-CoV-2

TL;DR

Ab-Affinity is introduced, a new large language model that can accurately predict the binding affinity of antibodies against a target peptide, e.g., the SARS-CoV-2 spike protein.

Abstract

Machine learning-based antibody design is emerging as one of the most promising approaches to combat infectious diseases, due to significant advancements in the field of artificial intelligence and an exponential surge in experimental antibody data (in particular related to COVID-19). The ability of an antibody to bind to an antigens (called binding affinity) is one of the the most critical properties in designing neutralizing antibodies. In this study we introduce Ab-Affinity, a new large language model that can accurately predict the binding affinity of antibodies against a target peptide, e.g., the SARS-CoV-2 spike protein. Code and model are available at https://github.com/ucrbioinfo/AbAffinity.
Paper Structure (13 sections, 8 figures, 1 table)

This paper contains 13 sections, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Dataset
  4. Model Architecture
  5. Training
  6. t-SNE visualization of embeddings
  7. Experimental Results
  8. Ab-Affinity Predicts the Binding Affinity of Antibodies Against SARS-CoV-2
  9. Ab-Affinity's Embeddings enable Downstream Classification Tasks
  10. Ab-Affinity's Attention Maps for Strong and Weak Binding Antibodies
  11. Ab-Affinity Captures Thermostabilty Properties of Antibodies
  12. Conclusion
  13. Acknowledgment

Figures (8)

  • Figure 1: Overview of Ab-Affinity. Ab-Affinity predicts the binding affinity against a specific target peptide; it can also provide residue-residue contact maps and an embedding of the input sequence.
  • Figure 2: Dataset description. a. Distribution of Binding Affinity ($\log~K_D$); b. Distribution of antibodies from each seed; c. Distribution of antibodies with mutation.
  • Figure 3: Model Architecture of Ab-Affinity
  • Figure 4: Comparing affinity prediction models: (a) Pearson and Spearman correlation for DG-Affinity (p-values = $3.86\times10^{-14}$, and $3.38\times10^{-15}$), ESM-2 (p-values = $8.03\times10^{-198}$, and $8.02\times10^{-198}$), AbLang (p-values = $1.24\times10^{-175}$, and $1.02\times10^{-163}$) and Ab-Affinity (p-values = $4.03\times10^{-261}$, and $8.65\times10^{-217}$); scatter plot for actual vs. predicted binding affinity for (b) DG-Affinity, (c) ESM-2, (d) AbLang, (e) Ab-Affinity (includes antibodies for all three seeds)
  • Figure 5: t-SNE representation of the embedding produced by ESM-2 and Ab-Affinity; antibodies are colored according to their predicted binding affinity.
  • ...and 3 more figures