Diagnosing genetic disorders requires extensive manual curation and interpretation of candidate variants, a labor-intensive task even for trained geneticists. Although artificial intelligence (AI) shows promise in aiding these diagnoses, existing AI tools have only achieved moderate success for primary diagnosis. AI-MARRVEL (AIM) uses a random-forest machine-learning classifier trained on over 3.5 million variants from thousands of diagnosed cases. AIM additionally incorporates expert-engineered features into training to recapitulate the intricate decision-making processes in molecular diagnosis. The online version of AIM is available at https://ai.marrvel.org. To evaluate AIM, we benchmarked it with diagnosed patients from three independent cohorts. AIM improved the rate of accurate genetic diagnosis, doubling the number of solved cases as compared with benchmarked methods, across three distinct real-world cohorts. To better identify diagnosable cases from the unsolved pools accumulated over time, we designed a confidence metric on which AIM achieved a precision rate of 98% and identified 57% of diagnosable cases out of a collection of 871 cases. Furthermore, AIM’s performance improved after being fine-tuned for targeted settings including recessive disorders and trio analysis. Finally, AIM demonstrated potential for novel disease gene discovery by correctly predicting two newly reported disease genes from the Undiagnosed Diseases Network. AIM achieved superior accuracy compared with existing methods for genetic diagnosis. We anticipate that this tool may aid in primary diagnosis, reanalysis of unsolved cases, and the discovery of novel disease genes. (Funded by the NIH Common Fund and others.)
Spatially resolved transcriptomics is a relatively new technique that maps transcriptional information within a tissue. Analysis of these datasets is challenging because gene expression values are highly sparse due to dropout events, and there is a lack of tools to facilitate in silico detection and annotation of regions based on their molecular content. Therefore, we develop a computational tool for detecting molecular regions and region-based Missing value Imputation for Spatially Transcriptomics (MIST). We validate MIST-identified regions across multiple datasets produced by 10x Visium Spatial Transcriptomics, using manually annotated histological images as references. We benchmark MIST against a spatial k-nearest neighboring baseline and other imputation methods designed for single-cell RNA sequencing. We use holdout experiments to demonstrate that MIST accurately recovers spatial transcriptomics missing values. MIST facilitates identifying intra-tissue heterogeneity and recovering spatial gene-gene co-expression signals. Using MIST before downstream analysis thus provides unbiased region detections to facilitate annotations with the associated functional analyses and produces accurately denoised spatial gene expression profiles.
Heterogeneous ensembles are an effective approach in scenarios where the ideal data type and/or individual predictor are unclear for a given problem. These ensembles have shown promise for protein function prediction (PFP), but their ability to improve PFP at a large scale is unclear. The overall goal of this study is to critically assess this ability of a variety of heterogeneous ensemble methods across a multitude of functional terms, proteins and organisms. Our results show that these methods, especially Stacking using Logistic Regression, indeed produce more accurate predictions for a variety of Gene Ontology terms differing in size and specificity. To enable the application of these methods to other related problems, we have publicly shared the HPC-enabled code underlying this work as LargeGOPred (https://github.com/GauravPandeyLab/LargeGOPred).
Summary: In the era where transcriptome profiling moves toward single-cell and spatial resolutions, the traditional co-expression analysis lacks the power to fully utilize such rich information to unravel spatial gene associations. Here, we present a Python package called Spatial Enrichment Analysis of Gene Associations using L-index (SEAGAL) to detect and visualize spatial gene correlations at both single-gene and gene-set levels. Our package takes spatial transcriptomics datasets with gene expression and the aligned spatial coordinates as input. It allows for analyzing and visualizing genes’ spatial correlations and cell types’ colocalization within the precise spatial context. The output could be visualized as volcano plots and heatmaps with a few lines of code, thus providing an easy-yet-comprehensive tool for mining spatial gene associations.Availability and implementation: The Python package SEAGAL can be installed using pip: https://pypi.org/project/seagal/. The source code and step-by-step tutorials are available at: https://github.com/linhuawang/SEAGAL.
Motivation: Integrating multimodal data represents an effective approach to predicting biomedical characteristics, such as protein functions and disease outcomes. However, existing data integration approaches do not sufficiently address the heterogeneous semantics of multimodal data. In particular, early and intermediate approaches that rely on a uniform integrated representation reinforce the consensus among the modalities, but may lose exclusive local information. The alternative late integration approach that can address this challenge has not been systematically studied for biomedical problems. Results: We propose Ensemble Integration (EI) as a novel systematic implementation of the late integration approach. EI infers local predictive models from the individual data modalities using appropriate algorithms, and uses effective heterogeneous ensemble algorithms to integrate these local models into a global predictive model. We also propose a novel interpretation method for EI models. We tested EI on the problems of predicting protein function from multimodal STRING data, and mortality due to COVID-19 from multimodal data in electronic health records. We found that EI accomplished its goal of producing significantly more accurate predictions than each individual modality. It also performed better than several established early integration methods for each of these problems. The interpretation of a representative EI model for COVID-19 mortality prediction identified several disease-relevant features, such as laboratory test (blood urea nitrogen (BUN) and calcium) and vital sign measurements (minimum oxygen saturation) and demographics (age). These results demonstrated the effectiveness of the EI framework for biomedical data integration and predictive modeling. Availability: Code and data are available at https://github.com/GauravPandeyLab/ensemble_integration . Contact: gaurav.pandey@mssm.edu
We are pleased to introduce a first-of-its-kind tool that combines in-silico region detection and missing value estimation for spatially resolved transcriptomics. Spatial transcriptomics by 10X Visium (ST) is a new technology used to dissect gene and cell spatial organization. Analyzing this new type of data has two main challenges: automatically annotating the major tissue regions and excessive zero values of gene expression due to high dropout rates. We developed a computational tool—MIST—that addresses both challenges by automatically identifying tissue regions and estimating missing gene-expression values for each detected region. We validated MIST detected regions across multiple datasets using manual annotation on the histological staining images as references. We also demonstrated that MIST can accurately recover ST’s missing values through hold-out experiments. Furthermore, we showed that MIST could identify intra-tissue heterogeneity and recover spatial gene-gene co-expression signals. We therefore strongly encourage using MIST before downstream ST analysis because it provides unbiased region annotations and enables accurately denoised spatial gene-expression profiles.