ABSTRACT
The incredible capabilities of generative artificial intelligence models have inevitably led to their application in the domain of drug discovery. Within this domain, the vastness of chemical space motivates the development of more efficient methods for identifying regions with molecules that exhibit desired characteristics. In this work, we present a computationally efficient active learning methodology and demonstrate its applicability to targeted molecular generation. When applied to c-Abl kinase, a protein with FDA-approved small-molecule inhibitors, the model learns to generate molecules similar to the inhibitors without prior knowledge of their existence and even reproduces two of them exactly. We also show that the methodology is effective for a protein without any commercially available small-molecule inhibitors, the HNH domain of the CRISPR-associated protein 9 (Cas9) enzyme. To facilitate implementation and reproducibility, we made all of our software available through the open-source ChemSpaceAL Python package.
Subject(s)
Artificial Intelligence , Problem-Based Learning , Reproducibility of Results , Software , Drug DiscoveryABSTRACT
The incredible capabilities of generative artificial intelligence models have inevitably led to their application in the domain of drug discovery. Within this domain, the vastness of chemical space motivates the development of more efficient methods for identifying regions with molecules that exhibit desired characteristics. In this work, we present a computationally efficient active learning methodology that requires evaluation of only a subset of the generated data in the constructed sample space to successfully align a generative model with respect to a specified objective. We demonstrate the applicability of this methodology to targeted molecular generation by fine-tuning a GPT-based molecular generator toward a protein with FDA-approved small-molecule inhibitors, c-Abl kinase. Remarkably, the model learns to generate molecules similar to the inhibitors without prior knowledge of their existence, and even reproduces two of them exactly. We also show that the methodology is effective for a protein without any commercially available small-molecule inhibitors, the HNH domain of the CRISPR-associated protein 9 (Cas9) enzyme. We believe that the inherent generality of this method ensures that it will remain applicable as the exciting field of in silico molecular generation evolves. To facilitate implementation and reproducibility, we have made all of our software available through the open-source ChemSpaceAL Python package.
ABSTRACT
Applying deep learning concepts from image detection and graph theory has greatly advanced protein-ligand binding affinity prediction, a challenge with enormous ramifications for both drug discovery and protein engineering. We build upon these advances by designing a novel deep learning architecture consisting of a 3-dimensional convolutional neural network utilizing channel-wise attention and two graph convolutional networks utilizing attention-based aggregation of node features. HAC-Net (Hybrid Attention-Based Convolutional Neural Network) obtains state-of-the-art results on the PDBbind v.2016 core set, the most widely recognized benchmark in the field. We extensively assess the generalizability of our model using multiple train-test splits, each of which maximizes differences between either protein structures, protein sequences, or ligand extended-connectivity fingerprints of complexes in the training and test sets. Furthermore, we perform 10-fold cross-validation with a similarity cutoff between SMILES strings of ligands in the training and test sets and also evaluate the performance of HAC-Net on lower-quality data. We envision that this model can be extended to a broad range of supervised learning problems related to structure-based biomolecular property prediction. All of our software is available as an open-source repository at https://github.com/gregory-kyro/HAC-Net/, and the HACNet Python package is available through PyPI.
Subject(s)
Neural Networks, Computer , Proteins , Ligands , Proteins/chemistry , Protein Binding , SoftwareABSTRACT
Influenza viruses remain a major public health threat. Seasonal influenza vaccination in humans primarily stimulates pre-existing memory B cells, which differentiate into a transient wave of circulating antibody-secreting plasmablasts1-3. This recall response contributes to 'original antigenic sin'-the selective increase of antibody species elicited by previous exposures to influenza virus antigens4. It remains unclear whether such vaccination can also induce germinal centre reactions in the draining lymph nodes, where diversification and maturation of recruited B cells can occur5. Here we used ultrasound-guided fine needle aspiration to serially sample the draining lymph nodes and investigate the dynamics and specificity of germinal centre B cell responses after influenza vaccination in humans. Germinal centre B cells that bind to influenza vaccine could be detected as early as one week after vaccination. In three out of eight participants, we detected vaccine-binding germinal centre B cells up to nine weeks after vaccination. Between 12% and 88% of the responding germinal centre B cell clones overlapped with B cells detected among early circulating plasmablasts. These shared B cell clones had high frequencies of somatic hypermutation and encoded broadly cross-reactive monoclonal antibodies. By contrast, vaccine-induced B cell clones detected only in the germinal centre compartment exhibited significantly lower frequencies of somatic hypermutation and predominantly encoded strain-specific monoclonal antibodies, which suggests a naive B cell origin. Some of these strain-specific monoclonal antibodies recognized epitopes that were not targeted by the early plasmablast response. Thus, influenza virus vaccination in humans can elicit a germinal centre reaction that recruits B cell clones that can target new epitopes, thereby broadening the spectrum of vaccine-induced protective antibodies.