Search for: All records

Machine learning models exhibit strong performance on datasets with abundant labeled samples. However, for tabular datasets with extremely high d-dimensional features but limited n samples (i.e. d ≫ n), machine learning models struggle to achieve strong performance due to the risk of overfitting. Here, our key insight is that there is often abundant, auxiliary domain information describing input features which can be structured as a heterogeneous knowledge graph (KG). We propose PLATO, a method that achieves strong performance on tabular data with d ≫ n by using an auxiliary KG describing input features to regularize a multilayer perceptron (MLP). In PLATO, each input feature corresponds to a node in the auxiliary KG. In the MLP’s first layer, each input feature also corresponds to a weight vector. PLATO is based on the inductive bias that two input features corresponding to similar nodes in the auxiliary KG should have similar weight vectors in the MLP’s first layer. PLATO captures this inductive bias by inferring the weight vector for each input feature from its corresponding node in the KG via a trainable message-passing function. Across 6 d ≫ n datasets, PLATO outperforms 13 state-of-the-art baselines by up to 10.19%. 

Graph Neural Networks (GNNs) are powerful machine learning prediction models on graph-structured data. However, GNNs lack rigorous uncertainty estimates, limiting their reliable deployment in settings where the cost of errors is significant. We propose conformalized GNN (CF-GNN), extending conformal prediction (CP) to graph-based models for guaranteed uncertainty estimates. Given an entity in the graph, CF-GNN produces a prediction set/interval that provably contains the true label with pre-defined coverage probability (e.g. 90%). We establish a permutation invariance condition that enables the validity of CP on graph data and provide an exact characterization of the test-time coverage. Besides valid coverage, it is crucial to reduce the prediction set size/interval length for practical use. We observe a key connection between non-conformity scores and network structures, which motivates us to develop a topology-aware output correction model that learns to update the prediction and produces more efficient prediction sets/intervals. Extensive experiments show that CF-GNN achieves any pre-defined target marginal coverage while significantly reducing the prediction set/interval size by up to 74% over the baselines. It also empirically achieves satisfactory conditional coverage over various raw and network features. 

Understanding cellular responses to genetic perturbation is central to numerous biomedical applications, from identifying genetic interactions involved in cancer to developing methods for regenerative medicine. However, the combinatorial explosion in the number of possible multigene perturbations severely limits experimental interrogation. Here, we present graph-enhanced gene activation and repression simulator (GEARS), a method that integrates deep learning with a knowledge graph of gene–gene relationships to predict transcriptional responses to both single and multigene perturbations using single-cell RNA-sequencing data from perturbational screens. GEARS is able to predict outcomes of perturbing combinations consisting of genes that were never experimentally perturbed. GEARS exhibited 40% higher precision than existing approaches in predicting four distinct genetic interaction subtypes in a combinatorial perturbation screen and identified the strongest interactions twice as well as prior approaches. Overall, GEARS can predict phenotypically distinct effects of multigene perturbations and thus guide the design of perturbational experiments.

 

Drug resistance poses a significant challenge in cancer treatment. Despite the initial effectiveness of therapies such as chemotherapy, targeted therapy and immunotherapy, many patients eventually develop resistance. To gain deep insights into the underlying mechanisms, single-cell profiling has been performed to interrogate drug resistance at cell level. Herein, we have built the DRMref database (https://ccsm.uth.edu/DRMref/) to provide comprehensive characterization of drug resistance using single-cell data from drug treatment settings. The current version of DRMref includes 42 single-cell datasets from 30 studies, covering 382 samples, 13 major cancer types, 26 cancer subtypes, 35 treatment regimens and 42 drugs. All datasets in DRMref are browsable and searchable, with detailed annotations provided. Meanwhile, DRMref includes analyses of cellular composition, intratumoral heterogeneity, epithelial–mesenchymal transition, cell–cell interaction and differentially expressed genes in resistant cells. Notably, DRMref investigates the drug resistance mechanisms (e.g. Aberration of Drug’s Therapeutic Target, Drug Inactivation by Structure Modification, etc.) in resistant cells. Additional enrichment analysis of hallmark/KEGG (Kyoto Encyclopedia of Genes and Genomes)/GO (Gene Ontology) pathways, as well as the identification of microRNA, motif and transcription factors involved in resistant cells, is provided in DRMref for user’s exploration. Overall, DRMref serves as a unique single-cell-based resource for studying drug resistance, drug combination therapy and discovering novel drug targets.

 

StemDriver is a comprehensive knowledgebase dedicated to the functional annotation of genes participating in the determination of hematopoietic stem cell fate, available at http://biomedbdc.wchscu.cn/StemDriver/. By utilizing single-cell RNA sequencing data, StemDriver has successfully assembled a comprehensive lineage map of hematopoiesis, capturing the entire continuum from the initial formation of hematopoietic stem cells to the fully developed mature cells. Extensive exploration and characterization were conducted on gene expression features corresponding to each lineage commitment. At the current version, StemDriver integrates data from 42 studies, encompassing a diverse range of 14 tissue types spanning from the embryonic phase to adulthood. In order to ensure uniformity and reliability, all data undergo a standardized pipeline, which includes quality data pre-processing, cell type annotation, differential gene expression analysis, identification of gene categories correlated with differentiation, analysis of highly variable genes along pseudo-time, and exploration of gene expression regulatory networks. In total, StemDriver assessed the function of 23 839 genes for human samples and 29 533 genes for mouse samples. Simultaneously, StemDriver also provided users with reference datasets and models for cell annotation. We believe that StemDriver will offer valuable assistance to research focused on cellular development and hematopoiesis.

 

Aging entails gradual functional decline influenced by interconnected factors. Multiple hallmarks proposed as common and conserved underlying denominators of aging on the molecular, cellular and systemic levels across multiple species. Thus, understanding the function of aging hallmarks and their relationships across species can facilitate the translation of anti-aging drug development from model organisms to humans. Here, we built AgeAnnoMO (https://relab.xidian.edu.cn/AgeAnnoMO/#/), a knowledgebase of multi-omics annotation for animal aging. AgeAnnoMO encompasses an extensive collection of 136 datasets from eight modalities, encompassing 8596 samples from 50 representative species, making it a comprehensive resource for aging and longevity research. AgeAnnoMO characterizes multiple aging regulators across species via multi-omics data, comprehensively annotating aging-related genes, proteins, metabolites, mitochondrial genes, microbiotas and age-specific TCR and BCR sequences tied to aging hallmarks for these species and tissues. AgeAnnoMO not only facilitates a deeper and more generalizable understanding of aging mechanisms, but also provides potential insights of the specificity across tissues and species in aging process, which is important to develop the effective anti-aging interventions for diverse populations. We anticipate that AgeAnnoMO will provide a valuable resource for comprehending and integrating the conserved driving hallmarks in aging biology and identifying the targetable biomarkers for aging research.

 

Abstract Molecular interaction networks are powerful resources for molecular discovery. They are increasingly used with machine learning methods to predict biologically meaningful interactions. While deep learning on graphs has dramatically advanced the prediction prowess, current graph neural network (GNN) methods are mainly optimized for prediction on the basis of direct similarity between interacting nodes. In biological networks, however, similarity between nodes that do not directly interact has proved incredibly useful in the last decade across a variety of interaction networks. Here, we present SkipGNN, a graph neural network approach for the prediction of molecular interactions. SkipGNN predicts molecular interactions by not only aggregating information from direct interactions but also from second-order interactions, which we call skip similarity. In contrast to existing GNNs, SkipGNN receives neural messages from two-hop neighbors as well as immediate neighbors in the interaction network and non-linearly transforms the messages to obtain useful information for prediction. To inject skip similarity into a GNN, we construct a modified version of the original network, called the skip graph. We then develop an iterative fusion scheme that optimizes a GNN using both the skip graph and the original graph. Experiments on four interaction networks, including drug–drug, drug–target, protein–protein, and gene–disease interactions, show that SkipGNN achieves superior and robust performance. Furthermore, we show that unlike popular GNNs, SkipGNN learns biologically meaningful embeddings and performs especially well on noisy, incomplete interaction networks. 

Abstract Summary Accurate prediction of drug–target interactions (DTI) is crucial for drug discovery. Recently, deep learning (DL) models for show promising performance for DTI prediction. However, these models can be difficult to use for both computer scientists entering the biomedical field and bioinformaticians with limited DL experience. We present DeepPurpose, a comprehensive and easy-to-use DL library for DTI prediction. DeepPurpose supports training of customized DTI prediction models by implementing 15 compound and protein encoders and over 50 neural architectures, along with providing many other useful features. We demonstrate state-of-the-art performance of DeepPurpose on several benchmark datasets. Availability and implementation https://github.com/kexinhuang12345/DeepPurpose. Supplementary information Supplementary data are available at Bioinformatics online. 

Therapeutics machine learning is an emerging field with incredible opportunities for innovatiaon and impact. However, advancement in this field requires formulation of meaningful learning tasks and careful curation of datasets. Here, we introduce Therapeutics Data Commons (TDC), the first unifying platform to systematically access and evaluate machine learning across the entire range of therapeutics. To date, TDC includes 66 AI-ready datasets spread across 22 learning tasks and spanning the discovery and development of safe and effective medicines. TDC also provides an ecosystem of tools and community resources, including 33 data functions and types of meaningful data splits, 23 strategies for systematic model evaluation, 17 molecule generation oracles, and 29 public leaderboards. All resources are integrated and accessible via an open Python library. We carry out extensive experiments on selected datasets, demonstrating that even the strongest algorithms fall short of solving key therapeutics challenges, including real dataset distributional shifts, multi-scale modeling of heterogeneous data, and robust generalization to novel data points. We envision that TDC can facilitate algorithmic and scientific advances and considerably accelerate machine-learning model development, validation and transition into biomedical and clinical implementation. TDC is an open-science initiative available at this https://tdcommons.ai.