ProDiGe: Prioritization Of Disease Genes with multitask machine learning from positive and unlabeled examples

Predicting disease-gene associations through self-supervised mutual infomax graph convolution network

Illuminating associations between diseases and genes can help reveal the pathogenesis of syndromes and contribute to treatments, but a large number of associations remained unexplored. To identify novel disease-gene associations, many computational methods have been developed using disease and gene-related prior knowledge. However, these methods remain of relatively inferior performance due to the limited external data sources and the inevitable noise among the prior knowledge. In this study, we have developed a new method, Self-Supervised Mutual Infomax Graph Convolution Network (MiGCN), to predict disease-gene associations under the guidance of external disease-disease and gene-gene collaborative graphs. The noises within the collaborative graphs were eliminated by maximizing the mutual information between nodes and neighbors through a graphical mutual infomax layer. In parallel, the node interactions were strengthened by a novel informative message passing layer to improve the learning ability of graph neural network. The extensive experiments showed that our model achieved performance improvement over the state-of-art method by more than 8 % on AUC. The datasets, source codes and trained models of MiGCN are available at https://github.com/biomed-AI/MiGCN.

Detecting biased validation of predictive models in the positive-unlabeled setting: disease gene prioritization case study

Motivation

Positive-unlabeled data consists of points with either positive or unknown labels. It is widespread in medical, genetic, and biological settings, creating a high demand for predictive positive-unlabeled models. The performance of such models is usually estimated using validation sets, assumed to be selected completely at random (SCAR) from known positive examples. For certain metrics, this assumption enables unbiased performance estimation when treating positive-unlabeled data as positive/negative. However, the SCAR assumption is often adopted without proper justifications, simply for the sake of convenience.

Results

We provide an algorithm that under the weak assumptions of a lower bound on the number of positive examples can test for the violation of the SCAR assumption. Applying it to the problem of gene prioritization for complex genetic traits, we illustrate that the SCAR assumption is often violated there, causing the inflation of performance estimates, which we refer to as validation bias. We estimate the potential impact of validation bias on performance estimation. Our analysis reveals that validation bias is widespread in gene prioritization data and can significantly overestimate the performance of models. This finding elucidates the discrepancy between the reported good performance of models and their limited practical applications.

Availability and implementation

Python code with examples of application of the validation bias detection algorithm is available at github.com/ArtomovLab/ValidationBias.

XGDAG: explainable gene-disease associations via graph neural networks

MOTIVATION

Disease gene prioritization consists in identifying genes that are likely to be involved in the mechanisms of a given disease, providing a ranking of such genes. Recently, the research community has used computational methods to uncover unknown gene-disease associations; these methods range from combinatorial to machine learning-based approaches. In particular, during the last years, approaches based on deep learning have provided superior results compared to more traditional ones. Yet, the problem with these is their inherent black-box structure, which prevents interpretability.

RESULTS

We propose a new methodology for disease gene discovery, which leverages graph-structured data using graph neural networks (GNNs) along with an explainability phase for determining the ranking of candidate genes and understanding the model's output. Our approach is based on a positive-unlabeled learning strategy, which outperforms existing gene discovery methods by exploiting GNNs in a non-black-box fashion. Our methodology is effective even in scenarios where a large number of associated genes need to be retrieved, in which gene prioritization methods often tend to lose their reliability.

AVAILABILITY AND IMPLEMENTATION

The source code of XGDAG is available on GitHub at: https://github.com/GiDeCarlo/XGDAG. The data underlying this article are available at: https://www.disgenet.org/, https://thebiogrid.org/, https://doi.org/10.1371/journal.pcbi.1004120.s003, and https://doi.org/10.1371/journal.pcbi.1004120.s004.

RelCurator: a text mining-based curation system for extracting gene-phenotype relationships specific to neurodegenerative disorders

BACKGROUND

The identification of gene-phenotype relationships is important in medical genetics as it serves as a basis for precision medicine. However, most of the gene-phenotype relationship data are buried in the biomedical literature in textual form.

OBJECTIVE

We propose RelCurator, a curation system that extracts sentences including both gene and phenotype entities related to specific disease categories from PubMed articles, provides rich additional information such as entity taggings, and predictions of gene-phenotype relationships.

METHODS

We targeted neurodegenerative disorders and developed a deep learning model using Bidirectional Gated Recurrent Unit (BiGRU) networks and BioWordVec word embeddings for predicting gene-phenotype relationships from biomedical texts. The prediction model is trained with more than 130,000 labeled PubMed sentences including gene and phenotype entities, which are related to or unrelated to neurodegenerative disorders.

RESULTS

We compared the performance of our deep learning model with those of Bidirectional Encoder Representations from Transformers (BERT), Support Vector Machine (SVM), and simple Recurrent Neural Network (simple RNN) models. Our model performed better with an F1-score of 0.96. Furthermore, the evaluation done using a few curation cases in the real scenario showed the effectiveness of our work. Therefore, we conclude that RelCurator can identify not only new causative genes, but also new genes associated with neurodegenerative disorders' phenotype.

CONCLUSION

RelCurator is a user-friendly method for accessing deep learning-based supporting information and a concise web interface to assist curators while browsing the PubMed articles. Our curation process represents an important and broadly applicable improvement to the state of the art for the curation of gene-phenotype relationships.

BAYESIAN ANALYSIS FOR IMBALANCED POSITIVE-UNLABELLED DIAGNOSIS CODES IN ELECTRONIC HEALTH RECORDS

With the increasing availability of electronic health records (EHR), significant progress has been made on developing predictive inference and algorithms by health data analysts and researchers. However, the EHR data are notoriously noisy due to missing and inaccurate inputs despite the information is abundant. One serious problem is that only a small portion of patients in the database has confirmatory diagnoses while many other patients remain undiagnosed because they did not comply with the recommended examinations. The phenomenon leads to a so-called positive-unlabelled situation and the labels are extremely imbalanced. In this paper, we propose a model-based approach to classify the unlabelled patients by using a Bayesian finite mixture model. We also discuss the label switching issue for the imbalanced data and propose a consensus Monte Carlo approach to address the imbalance issue and improve computational efficiency simultaneously. Simulation studies show that our proposed model-based approach outperforms existing positive-unlabelled learning algorithms. The proposed method is applied on the Cerner EHR for detecting diabetic retinopathy (DR) patients using laboratory measurements. With only 3% confirmatory diagnoses in the EHR database, we estimate the actual DR prevalence to be 25% which coincides with reported findings in the medical literature.

HetIG-PreDiG: A Heterogeneous Integrated Graph Model for Predicting Human Disease Genes based on gene expression

Graph analytical approaches permit identifying novel genes involved in complex diseases, but are limited by (i) inferring structural network similarity of connected gene nodes, ignoring potentially relevant unconnected nodes; (ii) using homogeneous graphs, missing gene-disease associations' complexity; (iii) relying on disease/gene-phenotype associations' similarities, involving highly incomplete data; (iv) using binary classification, with gene-disease edges as positive training samples, and non-associated gene and disease nodes as negative samples that may include currently unknown disease genes; or (v) reporting predicted novel associations without systematically evaluating their accuracy. Addressing these limitations, we develop the Heterogeneous Integrated Graph for Predicting Disease Genes (HetIG-PreDiG) model that includes gene-gene, gene-disease, and gene-tissue associations. We predict novel disease genes using low-dimensional representation of nodes accounting for network structure, and extending beyond network structure using the developed Gene-Disease Prioritization Score (GDPS) reflecting the degree of gene-disease association via gene co-expression data. For negative training samples, we select non-associated gene and disease nodes with lower GDPS that are less likely to be affiliated. We evaluate the developed model's success in predicting novel disease genes by analyzing the prediction probabilities of gene-disease associations. HetIG-PreDiG successfully predicts (Micro-F1 = 0.95) gene-disease associations, outperforming baseline models, and is validated using published literature, thus advancing our understanding of complex genetic diseases.

Prioritization of New Candidate Genes for Rare Genetic Diseases by a Disease-Aware Evaluation of Heterogeneous Molecular Networks

Screening for pathogenic variants in the diagnosis of rare genetic diseases can now be performed on all genes thanks to the application of whole exome and genome sequencing (WES, WGS). Yet the repertoire of gene-disease associations is not complete. Several computer-based algorithms and databases integrate distinct gene-gene functional networks to accelerate the discovery of gene-disease associations. We hypothesize that the ability of every type of information to extract relevant insights is disease-dependent. We compiled 33 functional networks classified into 13 knowledge categories (KCs) and observed large variability in their ability to recover genes associated with 91 genetic diseases, as measured using efficiency and exclusivity. We developed GLOWgenes, a network-based algorithm that applies random walk with restart to evaluate KCs' ability to recover genes from a given list associated with a phenotype and modulates the prediction of new candidates accordingly. Comparison with other integration strategies and tools shows that our disease-aware approach can boost the discovery of new gene-disease associations, especially for the less obvious ones. KC contribution also varies if obtained using recently discovered genes. Applied to 15 unsolved WES, GLOWgenes proposed three new genes to be involved in the phenotypes of patients with syndromic inherited retinal dystrophies.

Geographical classification of malaria parasites through applying machine learning to whole genome sequence data

Malaria, caused by Plasmodium parasites, is a major global health challenge. Whole genome sequencing (WGS) of Plasmodium falciparum and Plasmodium vivax genomes is providing insights into parasite genetic diversity, transmission patterns, and can inform decision making for clinical and surveillance purposes. Advances in sequencing technologies are helping to generate timely and big genomic datasets, with the prospect of applying Artificial Intelligence analytical techniques (e.g., machine learning) to support programmatic malaria control and elimination. Here, we assess the potential of applying deep learning convolutional neural network approaches to predict the geographic origin of infections (continents, countries, GPS locations) using WGS data of P. falciparum (n = 5957; 27 countries) and P. vivax (n = 659; 13 countries) isolates. Using identified high-quality genome-wide single nucleotide polymorphisms (SNPs) (P. falciparum: 750 k, P. vivax: 588 k), an analysis of population structure and ancestry revealed clustering at the country-level. When predicting locations for both species, classification (compared to regression) methods had the lowest distance errors, and > 90% accuracy at a country level. Our work demonstrates the utility of machine learning approaches for geo-classification of malaria parasites. With timelier WGS data generation across more malaria-affected regions, the performance of machine learning approaches for geo-classification will improve, thereby supporting disease control activities.

A gradient boosting tree model for multi-department venous thromboembolism risk assessment with imbalanced data

Venous thromboembolism (VTE) is the world's third most common cause of vascular mortality and a serious complication from multiple departments. Risk assessment of VTE guides clinical intervention in time and is of great importance to in-hospital patients. Traditional VTE risk assessment methods based on scaling tools, which always require rules carefully designed by human experts, are difficult to apply to large-population scenarios since the manually designed rules are not guaranteed to be accurate to all populations. In contrast, with the development of the electronic health record (EHR) datasets, data-driven machine-learning-based risk assessment methods have proven superior predictability in many studies in recent years. This paper uses the gradient boosting tree model to study the VTE risk assessment problem with multi-department data. There exist two distinct characteristics of VTE data collected at the level of the entire hospital: its wide distribution and heterogeneity across multiple departments. To this end, we consider the prediction task over multiple departments as a multi-task learning process, and introduce the algorithm of a task-aware tree-based method TSGB to tackle the multi-task prediction problem. Although the introduction of multi-task learning improves overall across-department performance, we reveal the problem of task-wise performance decline while dealing with imbalanced VTE data volume. According to the analysis, we finally propose two variants of TSGB to alleviate the problems and further boost the prediction performance. Compared with state-of-the-art rule-based and multi-task tree-based methods, the experimental results show the proposed methods not only improve the overall across-department AUC performance effectively, but also ensure the improvement of performance over every single department prediction.

Angiogenesis goes computational - The future way forward to discover new angiogenic targets?

Multi-omics technologies are being increasingly utilized in angiogenesis research. Yet, computational methods have not been widely used for angiogenic target discovery and prioritization in this field, partly because (wet-lab) vascular biologists are insufficiently familiar with computational biology tools and the opportunities they may offer. With this review, written for vascular biologists who lack expertise in computational methods, we aspire to break boundaries between both fields and to illustrate the potential of these tools for future angiogenic target discovery. We provide a comprehensive survey of currently available computational approaches that may be useful in prioritizing candidate genes, predicting associated mechanisms, and identifying their specificity to endothelial cell subtypes. We specifically highlight tools that use flexible, machine learning frameworks for large-scale data integration and gene prioritization. For each purpose-oriented category of tools, we describe underlying conceptual principles, highlight interesting applications and discuss limitations. Finally, we will discuss challenges and recommend some guidelines which can help to optimize the process of accurate target discovery.

Machine learning prediction and tau-based screening identifies potential Alzheimer's disease genes relevant to immunity

With increased research funding for Alzheimer's disease (AD) and related disorders across the globe, large amounts of data are being generated. Several studies employed machine learning methods to understand the ever-growing omics data to enhance early diagnosis, map complex disease networks, or uncover potential drug targets. We describe results based on a Target Central Resource Database protein knowledge graph and evidence paths transformed into vectors by metapath matching. We extracted features between specific genes and diseases, then trained and optimized our model using XGBoost, termed MPxgb(AD). To determine our MPxgb(AD) prediction performance, we examined the top twenty predicted genes through an experimental screening pipeline. Our analysis identified potential AD risk genes: FRRS1, CTRAM, SCGB3A1, FAM92B/CIBAR2, and TMEFF2. FRRS1 and FAM92B are considered dark genes, while CTRAM, SCGB3A1, and TMEFF2 are connected to TREM2-TYROBP, IL-1β-TNFα, and MTOR-APP AD-risk nodes, suggesting relevance to the pathogenesis of AD.

Machine learning and network medicine approaches for drug repositioning for COVID-19

We present two machine learning approaches for drug repurposing. While we have developed them for COVID-19, they are disease-agnostic. The two methodologies are complementary, targeting SARS-CoV-2 and host factors, respectively. Our first approach consists of a matrix factorization algorithm to rank broad-spectrum antivirals. Our second approach, based on network medicine, uses graph kernels to rank drugs according to the perturbation they induce on a subnetwork of the human interactome that is crucial for SARS-CoV-2 infection/replication. Our experiments show that our top predicted broad-spectrum antivirals include drugs indicated for compassionate use in COVID-19 patients; and that the ranking obtained by our kernel-based approach aligns with experimental data. Finally, we present the COVID-19 repositioning explorer (CoREx), an interactive online tool to explore the interplay between drugs and SARS-CoV-2 host proteins in the context of biological networks, protein function, drug clinical use, and Connectivity Map. CoREx is freely available at: https://paccanarolab.org/corex/.

PDGNet: Predicting Disease Genes Using a Deep Neural Network With Multi-View Features

The knowledge of phenotype-genotype associations is crucial for the understanding of disease mechanisms. Numerous studies have focused on developing efficient and accurate computing approaches to predict disease genes. However, owing to the sparseness and complexity of medical data, developing an efficient deep neural network model to identify disease genes remains a huge challenge. Therefore, we develop a novel deep neural network model that fuses the multi-view features of phenotypes and genotypes to identify disease genes (termed PDGNet). Our model integrated the multi-view features of diseases and genes and leveraged the feedback information of training samples to optimize the parameters of deep neural network and obtain the deep vector features of diseases and genes. The evaluation experiments on a large data set indicated that PDGNet obtained higher performance than the state-of-the-art method (precision and recall improved by 9.55 and 9.63 percent). The analysis results for the candidate genes indicated that the predicted genes have strong functional homogeneity and dense interactions with known genes. We validated the top predicted genes of Parkinson's disease based on external curated data and published medical literatures, which indicated that the candidate genes have a huge potential to guide the selection of causal genes in the 'wet experiment'. The source codes and the data of PDGNet are available at https://github.com/yangkuoone/PDGNet.

Positive-unlabeled learning in bioinformatics and computational biology: a brief review

Conventional supervised binary classification algorithms have been widely applied to address significant research questions using biological and biomedical data. This classification scheme requires two fully labeled classes of data (e.g. positive and negative samples) to train a classification model. However, in many bioinformatics applications, labeling data is laborious, and the negative samples might be potentially mislabeled due to the limited sensitivity of the experimental equipment. The positive unlabeled (PU) learning scheme was therefore proposed to enable the classifier to learn directly from limited positive samples and a large number of unlabeled samples (i.e. a mixture of positive or negative samples). To date, several PU learning algorithms have been developed to address various biological questions, such as sequence identification, functional site characterization and interaction prediction. In this paper, we revisit a collection of 29 state-of-the-art PU learning bioinformatic applications to address various biological questions. Various important aspects are extensively discussed, including PU learning methodology, biological application, classifier design and evaluation strategy. We also comment on the existing issues of PU learning and offer our perspectives for the future development of PU learning applications. We anticipate that our work serves as an instrumental guideline for a better understanding of the PU learning framework in bioinformatics and further developing next-generation PU learning frameworks for critical biological applications.

Identifying potential association on gene-disease network via dual hypergraph regularized least squares

BACKGROUND

Identifying potential associations between genes and diseases via biomedical experiments must be the time-consuming and expensive research works. The computational technologies based on machine learning models have been widely utilized to explore genetic information related to complex diseases. Importantly, the gene-disease association detection can be defined as the link prediction problem in bipartite network. However, many existing methods do not utilize multiple sources of biological information; Additionally, they do not extract higher-order relationships among genes and diseases.

RESULTS

In this study, we propose a novel method called Dual Hypergraph Regularized Least Squares (DHRLS) with Centered Kernel Alignment-based Multiple Kernel Learning (CKA-MKL), in order to detect all potential gene-disease associations. First, we construct multiple kernels based on various biological data sources in gene and disease spaces respectively. After that, we use CAK-MKL to obtain the optimal kernels in the two spaces respectively. To specific, hypergraph can be employed to establish higher-order relationships. Finally, our DHRLS model is solved by the Alternating Least squares algorithm (ALSA), for predicting gene-disease associations.

CONCLUSION

Comparing with many outstanding prediction tools, DHRLS achieves best performance on gene-disease associations network under two types of cross validation. To verify robustness, our proposed approach has excellent prediction performance on six real-world networks. Our research work can effectively discover potential disease-associated genes and provide guidance for the follow-up verification methods of complex diseases.

Using deep learning to identify recent positive selection in malaria parasite sequence data

BACKGROUND

Malaria, caused by Plasmodium parasites, is a major global public health problem. To assist an understanding of malaria pathogenesis, including drug resistance, there is a need for the timely detection of underlying genetic mutations and their spread. With the increasing use of whole-genome sequencing (WGS) of Plasmodium DNA, the potential of deep learning models to detect loci under recent positive selection, historically signals of drug resistance, was evaluated.

METHODS

A deep learning-based approach (called "DeepSweep") was developed, which can be trained on haplotypic images from genetic regions with known sweeps, to identify loci under positive selection. DeepSweep software is available from https://github.com/WDee/Deepsweep .

RESULTS

Using simulated genomic data, DeepSweep could detect recent sweeps with high predictive accuracy (areas under ROC curve > 0.95). DeepSweep was applied to Plasmodium falciparum (n = 1125; genome size 23 Mbp) and Plasmodium vivax (n = 368; genome size 29 Mbp) WGS data, and the genes identified overlapped with two established extended haplotype homozygosity methods (within-population iHS, across-population Rsb) (~ 60-75% overlap of hits at P < 0.0001). DeepSweep hits included regions proximal to known drug resistance loci for both P. falciparum (e.g. pfcrt, pfdhps and pfmdr1) and P. vivax (e.g. pvmrp1).

CONCLUSION

The deep learning approach can detect positive selection signatures in malaria parasite WGS data. Further, as the approach is generalizable, it may be trained to detect other types of selection. With the ability to rapidly generate WGS data at low cost, machine learning approaches (e.g. DeepSweep) have the potential to assist parasite genome-based surveillance and inform malaria control decision-making.

N-semble-based method for identifying Parkinson’s disease genes

Parkinson’s disease (PD) genes identification plays an important role in improving the diagnosis and treatment of the disease. A number of machine learning methods have been proposed to identify disease-related genes, but only few of these methods are adopted for PD. This work puts forth a novel neural network-based ensemble (n-semble) method to identify Parkinson’s disease genes. The artificial neural network is trained in a unique way to ensemble the multiple model predictions. The proposed n-semble method is composed of four parts: (1) protein sequences are used to construct feature vectors using physicochemical properties of amino acid; (2) dimensionality reduction is achieved using the t-Distributed Stochastic Neighbor Embedding (t-SNE) method, (3) the Jaccard method is applied to find likely negative samples from unknown (candidate) genes, and (4) gene prediction is performed with n-semble method. The proposed n-semble method has been compared with Smalter’s, ProDiGe, PUDI and EPU methods using various evaluation metrics. It has been concluded that the proposed n-semble method outperforms the existing gene identification methods over the other methods and achieves significantly higher precision, recall and F Score of 88.9%, 90.9% and 89.8%, respectively. The obtained results confirm the effectiveness and validity of the proposed framework.

Predicting disease‐associated genes: Computational methods, databases, and evaluations

Complex diseases are associated with a set of genes (called disease genes), the identification of which can help scientists uncover the mechanisms of diseases and develop new drugs and treatment strategies. Due to the huge cost and time of experimental identification techniques, many computational algorithms have been proposed to predict disease genes. Although several review publications in recent years have discussed many computational methods, some of them focus on cancer driver genes while others focus on biomolecular networks, which only cover a specific aspect of existing methods. In this review, we summarize existing methods and classify them into three categories based on their rationales. Then, the algorithms, biological data, and evaluation methods used in the computational prediction are discussed. Finally, we highlight the limitations of existing methods and point out some future directions for improving these algorithms. This review could help investigators understand the principles of existing methods, and thus develop new methods to advance the computational prediction of disease genes.

This article is categorized under:

A network-based machine-learning framework to identify both functional modules and disease genes

Disease gene identification is a critical step towards uncovering the molecular mechanisms of diseases and systematically investigating complex disease phenotypes. Despite considerable efforts to develop powerful computing methods, candidate gene identification remains a severe challenge owing to the connectivity of an incomplete interactome network, which hampers the discovery of true novel candidate genes. We developed a network-based machine-learning framework to identify both functional modules and disease candidate genes. In this framework, we designed a semi-supervised non-negative matrix factorization model to obtain the functional modules related to the diseases and genes. Of note, we proposed a disease gene-prioritizing method called MapGene that integrates the correlations from both functional modules and network closeness. Our framework identified a set of functional modules with highly functional homogeneity and close gene interactions. Experiments on a large-scale benchmark dataset showed that MapGene performs significantly better than the state-of-the-art algorithms. Further analysis demonstrates MapGene can effectively relieve the impact of the incompleteness of interactome networks and obtain highly reliable rankings of candidate genes. In addition, disease cases on Parkinson's disease and diabetes mellitus confirmed the generalization of MapGene for novel candidate gene identification. This work proposed, for the first time, an integrated computing framework to predict both functional modules and disease candidate genes. The methodology and results support that our framework has the potential to help discover underlying functional modules and reliable candidate genes in human disease.

Pathogenic gene prediction based on network embedding

In disease research, the study of gene-disease correlation has always been an important topic. With the emergence of large-scale connected data sets in biology, we use known correlations between the entities, which may be from different sets, to build a biological heterogeneous network and propose a new network embedded representation algorithm to calculate the correlation between disease and genes, using the correlation score to predict pathogenic genes. Then, we conduct several experiments to compare our method to other state-of-the-art methods. The results reveal that our method achieves better performance than the traditional methods.

Recent advances in network-based methods for disease gene prediction

Disease-gene association through genome-wide association study (GWAS) is an arduous task for researchers. Investigating single nucleotide polymorphisms that correlate with specific diseases needs statistical analysis of associations. Considering the huge number of possible mutations, in addition to its high cost, another important drawback of GWAS analysis is the large number of false positives. Thus, researchers search for more evidence to cross-check their results through different sources. To provide the researchers with alternative and complementary low-cost disease-gene association evidence, computational approaches come into play. Since molecular networks are able to capture complex interplay among molecules in diseases, they become one of the most extensively used data for disease-gene association prediction. In this survey, we aim to provide a comprehensive and up-to-date review of network-based methods for disease gene prediction. We also conduct an empirical analysis on 14 state-of-the-art methods. To summarize, we first elucidate the task definition for disease gene prediction. Secondly, we categorize existing network-based efforts into network diffusion methods, traditional machine learning methods with handcrafted graph features and graph representation learning methods. Thirdly, an empirical analysis is conducted to evaluate the performance of the selected methods across seven diseases. We also provide distinguishing findings about the discussed methods based on our empirical analysis. Finally, we highlight potential research directions for future studies on disease gene prediction.

Inferring Protein Sequence-Function Relationships with Large-Scale Positive-Unlabeled Learning

Machine learning can infer how protein sequence maps to function without requiring a detailed understanding of the underlying physical or biological mechanisms. It is challenging to apply existing supervised learning frameworks to large-scale experimental data generated by deep mutational scanning (DMS) and related methods. DMS data often contain high-dimensional and correlated sequence variables, experimental sampling error and bias, and the presence of missing data. Notably, most DMS data do not contain examples of negative sequences, making it challenging to directly estimate how sequence affects function. Here, we develop a positive-unlabeled (PU) learning framework to infer sequence-function relationships from large-scale DMS data. Our PU learning method displays excellent predictive performance across ten large-scale sequence-function datasets, representing proteins of different folds, functions, and library types. The estimated parameters pinpoint key residues that dictate protein structure and function. Finally, we apply our statistical sequence-function model to design highly stabilized enzymes.

MicroRNA-disease association prediction by matrix tri-factorization

Background

Biological evidence has shown that microRNAs(miRNAs) are greatly implicated in various biological progresses involved in human diseases. The identification of miRNA-disease associations(MDAs) is beneficial to disease diagnosis as well as treatment. Due to the high costs of biological experiments, it attracts more and more attention to predict MDAs by computational approaches.

Results

In this work, we propose a novel model MTFMDA for miRNA-disease association prediction by matrix tri-factorization, based on the known miRNA-disease associations, two types of miRNA similarities, and two types of disease similarities. The main idea of MTFMDA is to factorize the miRNA-disease association matrix to three matrices, a feature matrix for miRNAs, a feature matrix for diseases, and a low-rank relationship matrix. Our model incorporates the Laplacian regularizers which force the feature matrices to preserve the similarities of miRNAs or diseases. A novel algorithm is proposed to solve the optimization problem.

Conclusions

We evaluate our model by 5-fold cross validation by using known MDAs from HMDD V2.0 and show that our model could obtain the significantly highest AUCs among all the state-of-art methods. We further validate our method by applying it on colon and breast neoplasms in two different types of experiment settings. The new identified associated miRNAs for the two diseases could be verified by two other databases including dbDEMC and HMDD V3.0, which further shows the power of our proposed method.

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.

Deep Collaborative Filtering for Prediction of Disease Genes

Accurate prioritization of potential disease genes is a fundamental challenge in biomedical research. Various algorithms have been developed to solve such problems. Inductive Matrix Completion (IMC) is one of the most reliable models for its well-established framework and its superior performance in predicting gene-disease associations. However, the IMC method does not hierarchically extract deep features, which might limit the quality of recovery. In this case, the architecture of deep learning, which obtains high-level representations and handles noises and outliers presented in large-scale biological datasets, is introduced into the side information of genes in our Deep Collaborative Filtering (DCF) model. Further, for lack of negative examples, we also exploit Positive-Unlabeled (PU) learning formulation to low-rank matrix completion. Our approach achieves substantially improved performance over other state-of-the-art methods on diseases from the Online Mendelian Inheritance in Man (OMIM) database. Our approach is 10 percent more efficient than standard IMC in detecting a true association, and significantly outperforms other alternatives in terms of the precision-recall metric at the top-k predictions. Moreover, we also validate the disease with no previously known gene associations and newly reported OMIM associations. The experimental results show that DCF is still satisfactory for ranking novel disease phenotypes as well as mining unexplored relationships. The source code and the data are available at https://github.com/xzenglab/DCF.

On the Unreported-Profile-is-Negative Assumption for Predictive Cheminformatics

In cheminformatics, compound-target binding profiles has been a main source of data for research. For data repositories that only provide positive profiles, a popular assumption is that unreported profiles are all negative. In this paper, we caution the audience not to take this assumption for granted, and present empirical evidence of its ineffectiveness from a machine learning perspective. Our examination is based on a setting where binding profiles are used as features to train predictive models; we show (1) prediction performance degrades when the assumption fails and (2) explicit recovery of unreported profiles improves prediction performance. In particular, we propose a framework that jointly recovers profiles and learns predictive model, and show it achieves further performance improvement. The presented study not only suggests applying matrix recovery methods to recover unreported profiles, but also initiates a new missing feature problem which we called Learning with Positive and Unknown Features.

Machine learning-based approaches for disease gene prediction

Disease gene prediction is an essential issue in biomedical research. In the early days, annotation-based approaches were proposed for this problem. With the development of high-throughput technologies, interaction data between genes/proteins have grown quickly and covered almost genome and proteome; thus, network-based methods for the problem become prominent. In parallel, machine learning techniques, which formulate the problem as a classification, have also been proposed. Here, we firstly show a roadmap of the machine learning-based methods for the disease gene prediction. In the beginning, the problem was usually approached using a binary classification, where positive and negative training sample sets are comprised of disease genes and non-disease genes, respectively. The disease genes are ones known to be associated with diseases; meanwhile, non-disease genes were randomly selected from those not yet known to be associated with diseases. However, the later may contain unknown disease genes. To overcome this uncertainty of defining the non-disease genes, more realistic approaches have been proposed for the problem, such as unary and semi-supervised classification. Recently, more advanced methods, including ensemble learning, matrix factorization and deep learning, have been proposed for the problem. Secondly, 12 representative machine learning-based methods for the disease gene prediction were examined and compared in terms of prediction performance and running time. Finally, their advantages, disadvantages, interpretability and trust were also analyzed and discussed.

Reaching the End-Game for GWAS: Machine Learning Approaches for the Prioritization of Complex Disease Loci

Genome-wide association studies (GWAS) have revealed thousands of genetic loci that underpin the complex biology of many human traits. However, the strength of GWAS - the ability to detect genetic association by linkage disequilibrium (LD) - is also its limitation. Whilst the ever-increasing study size and improved design have augmented the power of GWAS to detect effects, differentiation of causal variants or genes from other highly correlated genes associated by LD remains the real challenge. This has severely hindered the biological insights and clinical translation of GWAS findings. Although thousands of disease susceptibility loci have been reported, causal genes at these loci remain elusive. Machine learning (ML) techniques offer an opportunity to dissect the heterogeneity of variant and gene signals in the post-GWAS analysis phase. ML models for GWAS prioritization vary greatly in their complexity, ranging from relatively simple logistic regression approaches to more complex ensemble models such as random forests and gradient boosting, as well as deep learning models, i.e., neural networks. Paired with functional validation, these methods show important promise for clinical translation, providing a strong evidence-based approach to direct post-GWAS research. However, as ML approaches continue to evolve to meet the challenge of causal gene identification, a critical assessment of the underlying methodologies and their applicability to the GWAS prioritization problem is needed. This review investigates the landscape of ML applications in three parts: selected models, input features, and output model performance, with a focus on prioritizations of complex disease associated loci. Overall, we explore the contributions ML has made towards reaching the GWAS end-game with consequent wide-ranging translational impact.

Pathogenic Gene Prediction Algorithm Based on Heterogeneous Information Fusion

Complex diseases seriously affect people's physical and mental health. The discovery of disease-causing genes has become a target of research. With the emergence of bioinformatics and the rapid development of biotechnology, to overcome the inherent difficulties of the long experimental period and high cost of traditional biomedical methods, researchers have proposed many gene prioritization algorithms that use a large amount of biological data to mine pathogenic genes. However, because the currently known gene-disease association matrix is still very sparse and lacks evidence that genes and diseases are unrelated, there are limits to the predictive performance of gene prioritization algorithms. Based on the hypothesis that functionally related gene mutations may lead to similar disease phenotypes, this paper proposes a PU induction matrix completion algorithm based on heterogeneous information fusion (PUIMCHIF) to predict candidate genes involved in the pathogenicity of human diseases. On the one hand, PUIMCHIF uses different compact feature learning methods to extract features of genes and diseases from multiple data sources, making up for the lack of sparse data. On the other hand, based on the prior knowledge that most of the unknown gene-disease associations are unrelated, we use the PU-Learning strategy to treat the unknown unlabeled data as negative examples for biased learning. The experimental results of the PUIMCHIF algorithm regarding the three indexes of precision, recall, and mean percentile ranking (MPR) were significantly better than those of other algorithms. In the top 100 global prediction analysis of multiple genes and multiple diseases, the probability of recovering true gene associations using PUIMCHIF reached 50% and the MPR value was 10.94%. The PUIMCHIF algorithm has higher priority than those from other methods, such as IMC and CATAPULT.

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.

HerGePred: Heterogeneous Network Embedding Representation for Disease Gene Prediction

The discovery of disease-causing genes is a critical step towards understanding the nature of a disease and determining a possible cure for it. In recent years, many computational methods to identify disease genes have been proposed. However, making full use of disease-related (e.g., symptoms) and gene-related (e.g., gene ontology and protein-protein interactions) information to improve the performance of disease gene prediction is still an issue. Here, we develop a heterogeneous disease-gene-related network (HDGN) embedding representation framework for disease gene prediction (called HerGePred). Based on this framework, a low-dimensional vector representation (LVR) of the nodes in the HDGN can be obtained. Then, we propose two specific algorithms, namely, an LVR-based similarity prediction and a random walk with restart on a reconstructed heterogeneous disease-gene network (RW-RDGN), to predict disease genes with high performance. First, to validate the rationality of the framework, we analyze the similarity-based overlap distribution of disease pairs and design an experiment for disease-gene association recovery, the results of which revealed that the LVR of nodes performs well at preserving the local and global network structure of the HDGN. Then, we apply tenfold cross validation and external validation to compare our methods with other well-known disease gene prediction algorithms. The experimental results show that the RW-RDGN performs better than the state-of-the-art algorithm. The prediction results of disease candidate genes are essential for molecular mechanism investigation and experimental validation. The source codes of HerGePred and experimental data are available at https://github.com/yangkuoone/HerGePred.

A novel one-class classification approach to accurately predict disease-gene association in acute myeloid leukemia cancer

Disease causing gene identification is considered as an important step towards drug design and drug discovery. In disease gene identification and classification, the main aim is to identify disease genes while identifying non-disease genes are of less or no significant. Hence, this task can be defined as a one-class classification problem. Existing machine learning methods typically take into consideration known disease genes as positive training set and unknown genes as negative samples to build a binary-class classification model. Here we propose a new One-class Classification Support Vector Machines (OCSVM) method to precisely classify candidate disease genes. Our aim is to build a model that concentrate its focus on detecting known disease-causing gene to increase sensitivity and precision. We investigate the impact of our proposed model using a benchmark consisting of the gene expression dataset for Acute Myeloid Leukemia (AML) cancer. Compared with the traditional methods, our experimental result shows the superiority of our proposed method in terms of precision, recall, and F-measure to detect disease causing genes for AML. OCSVM codes and our extracted AML benchmark are publicly available at: https://github.com/imandehzangi/OCSVM.

WMGHMDA: a novel weighted meta-graph-based model for predicting human microbe-disease association on heterogeneous information network

BACKGROUND

An increasing number of biological and clinical evidences have indicated that the microorganisms significantly get involved in the pathological mechanism of extensive varieties of complex human diseases. Inferring potential related microbes for diseases can not only promote disease prevention, diagnosis and treatment, but also provide valuable information for drug development. Considering that experimental methods are expensive and time-consuming, developing computational methods is an alternative choice. However, most of existing methods are biased towards well-characterized diseases and microbes. Furthermore, existing computational methods are limited in predicting potential microbes for new diseases.

RESULTS

Here, we developed a novel computational model to predict potential human microbe-disease associations (MDAs) based on Weighted Meta-Graph (WMGHMDA). We first constructed a heterogeneous information network (HIN) by combining the integrated microbe similarity network, the integrated disease similarity network and the known microbe-disease bipartite network. And then, we implemented iteratively pre-designed Weighted Meta-Graph search algorithm on the HIN to uncover possible microbe-disease pairs by cumulating the contribution values of weighted meta-graphs to the pairs as their probability scores. Depending on contribution potential, we described the contribution degree of different types of meta-graphs to a microbe-disease pair with bias rating. Meta-graph with higher bias rating will be assigned greater weight value when calculating probability scores.

CONCLUSIONS

The experimental results showed that WMGHMDA outperformed some state-of-the-art methods with average AUCs of 0.9288, 0.9068 ±0.0031 in global leave-one-out cross validation (LOOCV) and 5-fold cross validation (5-fold CV), respectively. In the case studies, 9, 19, 37 and 10, 20, 45 out of top-10, 20, 50 candidate microbes were manually verified by previous reports for asthma and inflammatory bowel disease (IBD), respectively. Furthermore, three common human diseases (Crohn's disease, Liver cirrhosis, Type 1 diabetes) were adopted to demonstrate that WMGHMDA could be efficiently applied to make predictions for new diseases. In summary, WMGHMDA has a high potential in predicting microbe-disease associations.

Predicting miRNA-disease association based on inductive matrix completion

Motivation

It has been shown that microRNAs (miRNAs) play key roles in variety of biological processes associated with human diseases. In Consideration of the cost and complexity of biological experiments, computational methods for predicting potential associations between miRNAs and diseases would be an effective complement.

Results

This paper presents a novel model of Inductive Matrix Completion for MiRNA-Disease Association prediction (IMCMDA). The integrated miRNA similarity and disease similarity are calculated based on miRNA functional similarity, disease semantic similarity and Gaussian interaction profile kernel similarity. The main idea is to complete the missing miRNA-disease association based on the known associations and the integrated miRNA similarity and disease similarity. IMCMDA achieves AUC of 0.8034 based on leave-one-out-cross-validation and improved previous models. In addition, IMCMDA was applied to five common human diseases in three types of case studies. In the first type, respectively, 42, 44, 45 out of top 50 predicted miRNAs of Colon Neoplasms, Kidney Neoplasms, Lymphoma were confirmed by experimental reports. In the second type of case study for new diseases without any known miRNAs, we chose Breast Neoplasms as the test example by hiding the association information between the miRNAs and Breast Neoplasms. As a result, 50 out of top 50 predicted Breast Neoplasms-related miRNAs are verified. In the third type of case study, IMCMDA was tested on HMDD V1.0 to assess the robustness of IMCMDA, 49 out of top 50 predicted Esophageal Neoplasms-related miRNAs are verified.

Availability and implementation

The code and dataset of IMCMDA are freely available at https://github.com/IMCMDAsourcecode/IMCMDA.

Supplementary information

Supplementary data are available at Bioinformatics online.

QTG-Finder: A Machine-Learning Based Algorithm To Prioritize Causal Genes of Quantitative Trait Loci in Arabidopsis and Rice

Linkage mapping is one of the most commonly used methods to identify genetic loci that determine a trait. However, the loci identified by linkage mapping may contain hundreds of candidate genes and require a time-consuming and labor-intensive fine mapping process to find the causal gene controlling the trait. With the availability of a rich assortment of genomic and functional genomic data, it is possible to develop a computational method to facilitate faster identification of causal genes. We developed QTG-Finder, a machine learning based algorithm to prioritize causal genes by ranking genes within a quantitative trait locus (QTL). Two predictive models were trained separately based on known causal genes in Arabidopsis and rice. An independent validation analysis showed that the models could recall about 64% of Arabidopsis and 79% of rice causal genes when the top 20% ranked genes were considered. The top 20% ranked genes can range from 10 to 100 genes, depending on the size of a QTL. The models can prioritize different types of traits though at different efficiency. We also identified several important features of causal genes including paralog copy number, being a transporter, being a transcription factor, and containing SNPs that cause premature stop codon. This work lays the foundation for systematically understanding characteristics of causal genes and establishes a pipeline to predict causal genes based on public data.

LOTUS: A single- and multitask machine learning algorithm for the prediction of cancer driver genes

Cancer driver genes, i.e., oncogenes and tumor suppressor genes, are involved in the acquisition of important functions in tumors, providing a selective growth advantage, allowing uncontrolled proliferation and avoiding apoptosis. It is therefore important to identify these driver genes, both for the fundamental understanding of cancer and to help finding new therapeutic targets or biomarkers. Although the most frequently mutated driver genes have been identified, it is believed that many more remain to be discovered, particularly for driver genes specific to some cancer types. In this paper, we propose a new computational method called LOTUS to predict new driver genes. LOTUS is a machine-learning based approach which allows to integrate various types of data in a versatile manner, including information about gene mutations and protein-protein interactions. In addition, LOTUS can predict cancer driver genes in a pan-cancer setting as well as for specific cancer types, using a multitask learning strategy to share information across cancer types. We empirically show that LOTUS outperforms five other state-of-the-art driver gene prediction methods, both in terms of intrinsic consistency and prediction accuracy, and provide predictions of new cancer genes across many cancer types.

Cancer development is driven by mutations and dysfunction of important, so-called cancer driver genes, that could be targeted by specific therapies. While a number of such cancer genes have already been identified, it is believed that many more remain to be discovered. To help prioritize experimental investigations of candidate genes, several computational methods have been proposed to rank promising candidates based on their mutations in large cohorts of cancer cases, or on their interactions with known driver genes in biological networks. We propose LOTUS, a new computational approach to identify genes with high oncogenic potential. LOTUS implements a machine learning approach to learn an oncogenic potential score from known driver genes, and brings two novelties compared to existing methods. First, it allows to easily combine heterogeneous sources of information into the scoring function, which we illustrate by learning a scoring function from both known mutations in large cancer cohorts and interactions in biological networks. Second, using a multitask learning strategy, it can predict different driver genes for different cancer types, while sharing information between them to improve the prediction for every type. We provide experimental results showing that LOTUS significantly outperforms several state-of-the-art cancer gene prediction software.

Predicting links between tumor samples and genes using 2-Layered graph based diffusion approach

Background

Determining the association between tumor sample and the gene is demanding because it requires a high cost for conducting genetic experiments. Thus, the discovered association between tumor sample and gene further requires clinical verification and validation. This entire mechanism is time-consuming and expensive. Due to this issue, predicting the association between tumor samples and genes remain a challenge in biomedicine.

Results

Here we present, a computational model based on a heat diffusion algorithm which can predict the association between tumor samples and genes. We proposed a 2-layered graph. In the first layer, we constructed a graph of tumor samples and genes where these two types of nodes are connected by “hasGene” relationship. In the second layer, the gene nodes are connected by “interaction” relationship. We applied the heat diffusion algorithms in nine different variants of genetic interaction networks extracted from STRING and BioGRID database. The heat diffusion algorithm predicted the links between tumor samples and genes with mean AUC-ROC score of 0.84. This score is obtained by using weighted genetic interactions of fusion or co-occurrence channels from the STRING database. For the unweighted genetic interaction from the BioGRID database, the algorithms predict the links with an AUC-ROC score of 0.74.

Conclusions

We demonstrate that the gene-gene interaction scores could improve the predictive power of the heat diffusion model to predict the links between tumor samples and genes. We showed the efficient runtime of the heat diffusion algorithm in various genetic interaction network. We statistically validated our prediction quality of the links between tumor samples and genes.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-3056-2) contains supplementary material, which is available to authorized users.

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.

Benchmarking network propagation methods for disease gene identification

In-silico identification of potential target genes for disease is an essential aspect of drug target discovery. Recent studies suggest that successful targets can be found through by leveraging genetic, genomic and protein interaction information. Here, we systematically tested the ability of 12 varied algorithms, based on network propagation, to identify genes that have been targeted by any drug, on gene-disease data from 22 common non-cancerous diseases in OpenTargets. We considered two biological networks, six performance metrics and compared two types of input gene-disease association scores. The impact of the design factors in performance was quantified through additive explanatory models. Standard cross-validation led to over-optimistic performance estimates due to the presence of protein complexes. In order to obtain realistic estimates, we introduced two novel protein complex-aware cross-validation schemes. When seeding biological networks with known drug targets, machine learning and diffusion-based methods found around 2-4 true targets within the top 20 suggestions. Seeding the networks with genes associated to disease by genetics decreased performance below 1 true hit on average. The use of a larger network, although noisier, improved overall performance. We conclude that diffusion-based prioritisers and machine learning applied to diffusion-based features are suited for drug discovery in practice and improve over simpler neighbour-voting methods. We also demonstrate the large impact of choosing an adequate validation strategy and the definition of seed disease genes.

The use of biological network data has proven its effectiveness in many areas from computational biology. Networks consist of nodes, usually genes or proteins, and edges that connect pairs of nodes, representing information such as physical interactions, regulatory roles or co-occurrence. In order to find new candidate nodes for a given biological property, the so-called network propagation algorithms start from the set of known nodes with that property and leverage the connections from the biological network to make predictions. Here, we assess the performance of several network propagation algorithms to find sensible gene targets for 22 common non-cancerous diseases, i.e. those that have been found promising enough to start the clinical trials with any compound. We focus on obtaining performance metrics that reflect a practical scenario in drug development where only a small set of genes can be essayed. We found that the presence of protein complexes biased the performance estimates, leading to over-optimistic conclusions, and introduced two novel strategies to address it. Our results support that network propagation is still a viable approach to find drug targets, but that special care needs to be put on the validation strategy. Algorithms benefitted from the use of a larger -although noisier- network and of direct evidence data, rather than indirect genetic associations to disease.

Gene prioritization using Bayesian matrix factorization with genomic and phenotypic side information

Motivation

Most gene prioritization methods model each disease or phenotype individually, but this fails to capture patterns common to several diseases or phenotypes. To overcome this limitation, we formulate the gene prioritization task as the factorization of a sparsely filled gene-phenotype matrix, where the objective is to predict the unknown matrix entries. To deliver more accurate gene-phenotype matrix completion, we extend classical Bayesian matrix factorization to work with multiple side information sources. The availability of side information allows us to make non-trivial predictions for genes for which no previous disease association is known.

Results

Our gene prioritization method can innovatively not only integrate data sources describing genes, but also data sources describing Human Phenotype Ontology terms. Experimental results on our benchmarks show that our proposed model can effectively improve accuracy over the well-established gene prioritization method, Endeavour. In particular, our proposed method offers promising results on diseases of the nervous system; diseases of the eye and adnexa; endocrine, nutritional and metabolic diseases; and congenital malformations, deformations and chromosomal abnormalities, when compared to Endeavour.

Availability and implementation

The Bayesian data fusion method is implemented as a Python/C++ package: https://github.com/jaak-s/macau. It is also available as a Julia package: https://github.com/jaak-s/BayesianDataFusion.jl. All data and benchmarks generated or analyzed during this study can be downloaded at https://owncloud.esat.kuleuven.be/index.php/s/UGb89WfkZwMYoTn.

Supplementary information

Supplementary data are available at Bioinformatics online.

Disease gene prediction for molecularly uncharacterized diseases

Network medicine approaches have been largely successful at increasing our knowledge of molecularly characterized diseases. Given a set of disease genes associated with a disease, neighbourhood-based methods and random walkers exploit the interactome allowing the prediction of further genes for that disease. In general, however, diseases with no known molecular basis constitute a challenge. Here we present a novel network approach to prioritize gene-disease associations that is able to also predict genes for diseases with no known molecular basis. Our method, which we have called Cardigan (ChARting DIsease Gene AssociatioNs), uses semi-supervised learning and exploits a measure of similarity between disease phenotypes. We evaluated its performance at predicting genes for both molecularly characterized and uncharacterized diseases in OMIM, using both weighted and binary interactomes, and compared it with state-of-the-art methods. Our tests, which use datasets collected at different points in time to replicate the dynamics of the disease gene discovery process, prove that Cardigan is able to accurately predict disease genes for molecularly uncharacterized diseases. Additionally, standard leave-one-out cross validation tests show how our approach outperforms state-of-the-art methods at predicting genes for molecularly characterized diseases by 14%-65%. Cardigan can also be used for disease module prediction, where it outperforms state-of-the-art methods by 87%-299%.