Predicting antibody binders and generating synthetic antibodies using deep learning

Article Biochemical Research Methods

DLAB: deep learning methods for structure-based virtual screening of antibodies

Constantin Schneider et al.

Summary: DLAB can be used to improve antibody-antigen docking and structure-based virtual screening, enhancing pose ranking for antibody docking experiments and selection of accurate and correctly ranked antibody-antigen pairings.

BIOINFORMATICS (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Antibody structure prediction using interpretable deep learning

Jeffrey A. Ruffolo et al.

Summary: DeepAb is a deep learning method that accurately predicts antibody FV structures from sequences. It outperforms alternative methods on diverse and therapeutically relevant antibodies and offers interpretable predictions by introducing an attention mechanism. Additionally, a mutant scoring metric derived from network confidence improves binding affinity for a specific antibody.

PATTERNS (2022)

添加到收藏夹

Article Multidisciplinary Sciences

DeepTCR is a deep learning framework for revealing sequence concepts within T-cell repertoires

John-William Sidhom et al.

Summary: Deep learning algorithms have been used to model complex TCR sequencing data and extract valuable information for both descriptive and predictive purposes. The results demonstrate the flexibility and capacity of deep neural networks in extracting meaningful information from complex immunogenomic data.

NATURE COMMUNICATIONS (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Antibody design using LSTM based deep generative model from phage display library for affinity maturation

Koichiro Saka et al.

Summary: This study presents an efficient method for discovering antibody sequences with higher affinity using a deep generative model LSTM. By applying this method to affinity maturation of antibodies against kynurenine, enriched sequences with significantly higher affinity were obtained. The trained LSTM model generated sequences with over 1800-fold higher affinity compared to the parental clone, demonstrating its effectiveness in affinity maturation.

SCIENTIFIC REPORTS (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Potential neutralizing antibodies discovered for novel corona virus using machine learning

Rishikesh Magar et al.

Summary: Researchers used a variety of machine learning models to predict synthetic antibodies that can inhibit SARS-CoV-2, and identified nine stable antibodies that potentially inhibit the virus. They combined bioinformatics, structural biology, and Molecular Dynamics simulations to verify the stability of these candidate antibodies.

SCIENTIFIC REPORTS (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Protein design and variant prediction using autoregressive generative models

Jung-Eun Shin et al.

Summary: The authors introduce a deep generative alignment-free model for sequence design applied to highly variable regions and design and test a diverse nanobody library with improved properties for selection experiments.

NATURE COMMUNICATIONS (2021)

添加到收藏夹

Article Engineering, Biomedical

Optimization of therapeutic antibodies by predicting antigen specificity from antibody sequence via deep learning

Derek M. Mason et al.

Summary: The optimization of therapeutic antibodies can be achieved by predicting antigen specificity using deep learning, resulting in the identification of optimized antibody variants. This approach may facilitate antibody engineering and optimization.

NATURE BIOMEDICAL ENGINEERING (2021)

添加到收藏夹

Article Biochemical Research Methods

Humanization of antibodies using a machine learning approach on large-scale repertoire data

Claire Marks et al.

Summary: The study developed machine learning classifiers to distinguish between human and non-human antibody sequences, and used them to create a computational humanization tool, Hu-mAb, which suggests mutations to reduce immunogenicity of antibody therapeutics. The tool's results show a negative relationship with experimental immunogenicity and can effectively replace trial-and-error humanization experiments in a shorter amount of time.

BIOINFORMATICS (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Highly accurate protein structure prediction with AlphaFold

John Jumper et al.

Summary: Proteins are essential for life, and accurate prediction of their structures is a crucial research problem. Current experimental methods are time-consuming, highlighting the need for accurate computational approaches to address the gap in structural coverage. Despite recent progress, existing methods fall short of atomic accuracy in protein structure prediction.

NATURE (2021)

添加到收藏夹

Article Multidisciplinary Sciences

Accurate prediction of protein structures and interactions using a three-track neural network

Minkyung Baek et al.

Summary: Through the three-track network, we achieved accuracies approaching those of DeepMind in CASP14, enabling rapid solution of challenging x-ray crystallography and cryo-electron microscopy structure modeling problems, and providing insights into the functions of proteins with currently unknown structure.

SCIENCE (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Machine Learning Detects Anti-DENV Signatures in Antibody Repertoire Sequences

Alexander Horst et al.

Summary: Dengue infection is a global threat without universal treatment or vaccines recommended by the WHO. Research on specific immune responses to the virus could aid in antibody discovery and vaccine design. High-throughput sequencing can identify antibodies produced in response to infection at the sequence level.

FRONTIERS IN ARTIFICIAL INTELLIGENCE (2021)

添加到收藏夹

Article Biochemical Research Methods