Scuba: scalable kernel-based gene prioritization

A publication-wide association study (PWAS), historical language models to prioritise novel therapeutic drug targets

Most biomedical knowledge is published as text, making it challenging to analyse using traditional statistical methods. In contrast, machine-interpretable data primarily comes from structured property databases, which represent only a fraction of the knowledge present in the biomedical literature. Crucial insights and inferences can be drawn from these publications by the scientific community. We trained language models on literature from different time periods to evaluate their ranking of prospective gene-disease associations and protein-protein interactions. Using 28 distinct historical text corpora of abstracts published between 1995 and 2022, we trained independent Word2Vec models to prioritise associations that were likely to be reported in future years. This study demonstrates that biomedical knowledge can be encoded as word embeddings without the need for human labelling or supervision. Language models effectively capture drug discovery concepts such as clinical tractability, disease associations, and biochemical pathways. Additionally, these models can prioritise hypotheses years before their initial reporting. Our findings underscore the potential for extracting yet-to-be-discovered relationships through data-driven approaches, leading to generalised biomedical literature mining for potential therapeutic drug targets. The Publication-Wide Association Study (PWAS) enables the prioritisation of under-explored targets and provides a scalable system for accelerating early-stage target ranking, irrespective of the specific disease of interest.

Predicting disease genes based on multi-head attention fusion

BACKGROUND

The identification of disease-related genes is of great significance for the diagnosis and treatment of human disease. Most studies have focused on developing efficient and accurate computational methods to predict disease-causing genes. Due to the sparsity and complexity of biomedical data, it is still a challenge to develop an effective multi-feature fusion model to identify disease genes.

RESULTS

This paper proposes an approach to predict the pathogenic gene based on multi-head attention fusion (MHAGP). Firstly, the heterogeneous biological information networks of disease genes are constructed by integrating multiple biomedical knowledge databases. Secondly, two graph representation learning algorithms are used to capture the feature vectors of gene-disease pairs from the network, and the features are fused by introducing multi-head attention. Finally, multi-layer perceptron model is used to predict the gene-disease association.

CONCLUSIONS

The MHAGP model outperforms all of other methods in comparative experiments. Case studies also show that MHAGP is able to predict genes potentially associated with diseases. In the future, more biological entity association data, such as gene-drug, disease phenotype-gene ontology and so on, can be added to expand the information in heterogeneous biological networks and achieve more accurate predictions. In addition, MHAGP with strong expansibility can be used for potential tasks such as gene-drug association and drug-disease association prediction.

Assignment of structural domains in proteins using diffusion kernels on graphs

Though proposing algorithmic approaches for protein domain decomposition has been of high interest, the inherent ambiguity to the problem makes it still an active area of research. Besides, accurate automated methods are in high demand as the number of solved structures for complex proteins is on the rise. While majority of the previous efforts for decomposition of 3D structures are centered on the developing clustering algorithms, employing enhanced measures of proximity between the amino acids has remained rather uncharted. If there exists a kernel function that in its reproducing kernel Hilbert space, structural domains of proteins become well separated, then protein structures can be parsed into domains without the need to use a complex clustering algorithm. Inspired by this idea, we developed a protein domain decomposition method based on diffusion kernels on protein graphs. We examined all combinations of four graph node kernels and two clustering algorithms to investigate their capability to decompose protein structures. The proposed method is tested on five of the most commonly used benchmark datasets for protein domain assignment plus a comprehensive non-redundant dataset. The results show a competitive performance of the method utilizing one of the diffusion kernels compared to four of the best automatic methods. Our method is also able to offer alternative partitionings for the same structure which is in line with the subjective definition of protein domain. With a competitive accuracy and balanced performance for the simple and complex structures despite relying on a relatively naive criterion to choose optimal decomposition, the proposed method revealed that diffusion kernels on graphs in particular, and kernel functions in general are promising measures to facilitate parsing proteins into domains and performing different structural analysis on proteins. The size and interconnectedness of the protein graphs make them promising targets for diffusion kernels as measures of affinity between amino acids. The versatility of our method allows the implementation of future kernels with higher performance. The source code of the proposed method is accessible at https://github.com/taherimo/kludo . Also, the proposed method is available as a web application from https://cbph.ir/tools/kludo .

Network-Based Approaches for Disease-Gene Association Prediction Using Protein-Protein Interaction Networks

Genome-wide association studies (GWAS) can be used to infer genome intervals that are involved in genetic diseases. However, investigating a large number of putative mutations for GWAS is resource- and time-intensive. Network-based computational approaches are being used for efficient disease-gene association prediction. Network-based methods are based on the underlying assumption that the genes causing the same diseases are located close to each other in a molecular network, such as a protein-protein interaction (PPI) network. In this survey, we provide an overview of network-based disease-gene association prediction methods based on three categories: graph-theoretic algorithms, machine learning algorithms, and an integration of these two. We experimented with six selected methods to compare their prediction performance using a heterogeneous network constructed by combining a genome-wide weighted PPI network, an ontology-based disease network, and disease-gene associations. The experiment was conducted in two different settings according to the presence and absence of known disease-associated genes. The results revealed that HerGePred, an integrative method, outperformed in the presence of known disease-associated genes, whereas PRINCE, which adopted a network propagation algorithm, was the most competitive in the absence of known disease-associated genes. Overall, the results demonstrated that the integrative methods performed better than the methods using graph-theory only, and the methods using a heterogeneous network performed better than those using a homogeneous PPI network only.

Knowledge graph-based recommendation framework identifies drivers of resistance in EGFR mutant non-small cell lung cancer

Resistance to EGFR inhibitors (EGFRi) presents a major obstacle in treating non-small cell lung cancer (NSCLC). One of the most exciting new ways to find potential resistance markers involves running functional genetic screens, such as CRISPR, followed by manual triage of significantly enriched genes. This triage process to identify 'high value' hits resulting from the CRISPR screen involves manual curation that requires specialized knowledge and can take even experts several months to comprehensively complete. To find key drivers of resistance faster we build a recommendation system on top of a heterogeneous biomedical knowledge graph integrating pre-clinical, clinical, and literature evidence. The recommender system ranks genes based on trade-offs between diverse types of evidence linking them to potential mechanisms of EGFRi resistance. This unbiased approach identifies 57 resistance markers from >3,000 genes, reducing hit identification time from months to minutes. In addition to reproducing known resistance markers, our method identifies previously unexplored resistance mechanisms that we prospectively validate.

Preclinical validation of therapeutic targets predicted by tensor factorization on heterogeneous graphs

Incorrect drug target identification is a major obstacle in drug discovery. Only 15% of drugs advance from Phase II to approval, with ineffective targets accounting for over 50% of these failures1-3. Advances in data fusion and computational modeling have independently progressed towards addressing this issue. Here, we capitalize on both these approaches with Rosalind, a comprehensive gene prioritization method that combines heterogeneous knowledge graph construction with relational inference via tensor factorization to accurately predict disease-gene links. Rosalind demonstrates an increase in performance of 18%-50% over five comparable state-of-the-art algorithms. On historical data, Rosalind prospectively identifies 1 in 4 therapeutic relationships eventually proven true. Beyond efficacy, Rosalind is able to accurately predict clinical trial successes (75% recall at rank 200) and distinguish likely failures (74% recall at rank 200). Lastly, Rosalind predictions were experimentally tested in a patient-derived in-vitro assay for Rheumatoid arthritis (RA), which yielded 5 promising genes, one of which is unexplored in RA.

Gene-set Enrichment with Mathematical Biology (GEMB)

BACKGROUND

Gene-set analyses measure the association between a disease of interest and a "set" of genes related to a biological pathway. These analyses often incorporate gene network properties to account for differential contributions of each gene. We extend this concept further-defining gene contributions based on biophysical properties-by leveraging mathematical models of biology to predict the effects of genetic perturbations on a particular downstream function.

RESULTS

We present a method that combines gene weights from model predictions and gene ranks from genome-wide association studies into a weighted gene-set test. We demonstrate in simulation how such a method can improve statistical power. To this effect, we identify a gene set, weighted by model-predicted contributions to intracellular calcium ion concentration, that is significantly related to bipolar disorder in a small dataset (P = 0.04; n = 544). We reproduce this finding using publicly available summary data from the Psychiatric Genomics Consortium (P = 1.7 × 10-4; n = 41,653). By contrast, an approach using a general calcium signaling pathway did not detect a significant association with bipolar disorder (P = 0.08). The weighted gene-set approach based on intracellular calcium ion concentration did not detect a significant relationship with schizophrenia (P = 0.09; n = 65,967) or major depression disorder (P = 0.30; n = 500,199).

CONCLUSIONS

Together, these findings show how incorporating math biology into gene-set analyses might help to identify biological functions that underlie certain polygenic disorders.

Heterogeneous networks integration for disease–gene prioritization with node kernels

Motivation

The identification of disease–gene associations is a task of fundamental importance in human health research. A typical approach consists in first encoding large gene/protein relational datasets as networks due to the natural and intuitive property of graphs for representing objects’ relationships and then utilizing graph-based techniques to prioritize genes for successive low-throughput validation assays. Since different types of interactions between genes yield distinct gene networks, there is the need to integrate different heterogeneous sources to improve the reliability of prioritization systems.

Results

We propose an approach based on three phases: first, we merge all sources in a single network, then we partition the integrated network according to edge density introducing a notion of edge type to distinguish the parts and finally, we employ a novel node kernel suitable for graphs with typed edges. We show how the node kernel can generate a large number of discriminative features that can be efficiently processed by linear regularized machine learning classifiers. We report state-of-the-art results on 12 disease–gene associations and on a time-stamped benchmark containing 42 newly discovered associations.

Availability and implementation

Source code: https://github.com/dinhinfotech/DiGI.git.

Supplementary information

Supplementary data are available at Bioinformatics online.

Genetic Analyses in Dent Disease and Characterization of CLCN5 Mutations in Kidney Biopsies

Dent disease (DD), an X-linked renal tubulopathy, is mainly caused by loss-of-function mutations in CLCN5 (DD1) and OCRL genes. CLCN5 encodes the ClC-5 antiporter that in proximal tubules (PT) participates in the receptor-mediated endocytosis of low molecular weight proteins. Few studies have analyzed the PT expression of ClC-5 and of megalin and cubilin receptors in DD1 kidney biopsies. About 25% of DD cases lack mutations in either CLCN5 or OCRL genes (DD3), and no other disease genes have been discovered so far. Sanger sequencing was used for CLCN5 gene analysis in 158 unrelated males clinically suspected of having DD. The tubular expression of ClC-5, megalin, and cubilin was assessed by immunolabeling in 10 DD1 kidney biopsies. Whole exome sequencing (WES) was performed in eight DD3 patients. Twenty-three novel CLCN5 mutations were identified. ClC-5, megalin, and cubilin were significantly lower in DD1 than in control biopsies. The tubular expression of ClC-5 when detected was irrespective of the type of mutation. In four DD3 patients, WES revealed 12 potentially pathogenic variants in three novel genes (SLC17A1, SLC9A3, and PDZK1), and in three genes known to be associated with monogenic forms of renal proximal tubulopathies (SLC3A, LRP2, and CUBN). The supposed third Dent disease-causing gene was not discovered.

A Survey of Gene Prioritization Tools for Mendelian and Complex Human Diseases

Modern high-throughput experiments provide us with numerous potential associations between genes and diseases. Experimental validation of all the discovered associations, let alone all the possible interactions between them, is time-consuming and expensive. To facilitate the discovery of causative genes, various approaches for prioritization of genes according to their relevance for a given disease have been developed. In this article, we explain the gene prioritization problem and provide an overview of computational tools for gene prioritization. Among about a hundred of published gene prioritization tools, we select and briefly describe 14 most up-to-date and user-friendly. Also, we discuss the advantages and disadvantages of existing tools, challenges of their validation, and the directions for future research.