14 June 2019 in PLOS Computational Biology
Identification of drug-target interactions (DTIs) plays a key role in drug discovery. The high cost and labor-intensive nature of in vitro and in vivo experiments have highlighted the importance of in silico-based DTI prediction approaches. In several computational models, conventional protein descriptors have been shown to not be sufficiently informative to predict accurate DTIs. Thus, in this study, we propose a deep learning based DTI prediction model capturing local residue patterns of proteins participating in DTIs. When we employ a convolutional neural network (CNN) on raw protein sequences, we perform convolution on various lengths of amino acids subsequences to capture local residue patterns of generalized protein classes. We train our model with large-scale DTI information and demonstrate the performance of the proposed model using an independent dataset that is not seen during the training phase. As a result, our model performs better than previous protein descriptor-based models. Also, our model performs better than the recently developed deep learning models for massive prediction of DTIs. By examining pooled convolution results, we confirmed that our model can detect binding sites of proteins for DTIs. In conclusion, our prediction model for detecting local residue patterns of target proteins successfully enriches the protein features of a raw protein sequence, yielding better prediction results than previous approaches. Our code is available at https://github.com/GIST-CSBL/DeepConv-DTI. Drugs work by interacting with target proteins to activate or inhibit a target’s biological process. Therefore, identification of DTIs is a crucial step in drug discovery. However, identifying drug candidates via biological assays is very time and cost consuming, which introduces the need for a computational prediction approach for the identification of DTIs. In this work, we constructed a novel DTI prediction model to extract local residue patterns of target protein sequences using a CNN-based deep learning approach. As a result, the detected local features of protein sequences perform better than other protein descriptors for DTI prediction and previous models for predicting PubChem independent test datasets. That is, our approach of capturing local residue patterns with CNN successfully enriches protein features from a raw sequence.
28 November 2024 in Nature Methods
24 October 2024 in bioRxiv
Genome-wide association studies have linked millions of genetic variants to biomedical phenotypes, but their utility has been limited by lack of mechanistic understanding and widespread epistatic interactions. Recently, Transformer models have emerged as powerful machine learning architectures with potential to address these and other challenges. Here we introduce the Genotype-to-Phenotype Transformer (G2PT), a framework for modeling hierarchical information flow among variants, genes, multigenic systems, and phenotypes. As proof-of-concept, we train G2PT to model the genetics of metabolic traits including insulin resistance (serum triglycerides-to-HDL ratio), LDL and type-2 diabetes. G2PT predicts these traits with accuracy exceeding state-of-the-art and, unlike other polygenic models, extends to distinct populations not used for training. Predictions of insulin resistance are based on >1,395 variants within 20 systems and include epistatic interactions among variants, e.g. between APOA4 and CETP in phospholipid transfer. This work positions hierarchical graph transformers as a next-generation approach to polygenic risk.
26 May 2026 in Cancer Discovery
While genetic sequencing is routine in cancer care, translating a tumor’s complex mutation profile into actionable treatment decisions remains a central challenge. MutationProjector is pre-trained from a large corpus of genomic alterations across 30,000+ tumors, integrated with extensive molecular knowledge. The resulting projection reveals a tumor’s altered molecular pathways, facilitating model interpretation, and it accurately reconstructs held-out mutations, demonstrating model generalization. When applied to predict immunotherapy or chemotherapy resistance across multiple cancer types and cohorts, MutationProjector achieves or exceeds state-of-the-art performance in all contexts. It identifies unexpected biomarkers, including KMT2D mutation in immunotherapy sensitivity and joint alteration of SMARCA4 and STK11 in immunotherapy resistance. These results establish a unifying framework for connecting tumor genotypes to biological mechanisms and therapeutic outcomes.