Search

The study of genetic variants can help find correlating population groups to identify cohorts that are predisposed to common diseases and explain differences in disease susceptibility and how patients react to drugs. Machine learning algorithms are increasingly being applied to identify interacting GVs to understand their complex phenotypic traits. Since the performance of a learning algorithm not only depends on the size and nature of the data but also on the quality of underlying representation, deep neural networks can learn non-linear mappings that allow transforming GVs data into more clustering and classification friendly representations than manual feature selection. In this paper, we proposed convolutional embedded networks in which we combine two DNN architectures called convolutional embedded clustering and convolutional autoencoder classifier for clustering individuals and predicting geographic ethnicity based on GVs, respectively. We employed CAE-based representation learning on 95 million GVs from the 1000 genomes and Simons genome diversity projects. Quantitative and qualitative analyses with a focus on accuracy and scalability show that our approach outperforms state-of-the-art approaches such as VariantSpark and ADMIXTURE. In particular, CEC can cluster targeted population groups in 22 hours with an adjusted rand index of 0.915, the normalized mutual information of 0.92, and the clustering accuracy of 89%. Contrarily, the CAE classifier can predict the geographic ethnicity of unknown samples with an F1 and Mathews correlation coefficient(MCC) score of 0.9004 and 0.8245, respectively. To provide interpretations of the predictions, we identify significant biomarkers using gradient boosted trees(GBT) and SHAP. Overall, our approach is transparent and faster than the baseline methods, and scalable for 5% to 100% of the full human genome., This article is under review in IEEE/ACM Transactions on Computational Biology and Bioinformatics. It is based on a workshop paper discussed at the Extended Semantic Web Conference (ESWC'2017) workshop on Semantic Web Solutions for Large-scale Biomedical Data Analytics (SeWeBMeDA), Slovenia, May, 28-29, 2017

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.

Prediction of metastatic sites from the primary site of origin is a impugn task in breast cancer (BRCA). Multi-dimensionality of such metastatic sites - bone, lung, kidney, and brain, using large-scale multi-dimensional Poly-Omics (Transcriptomics, Proteomics and Metabolomics) data of various type, for example, CNV (Copy number variation), GE (Gene expression), DNA methylation, path-ways, and drugs with clinical associations makes classification of metastasis a multi-faceted challenge. In this paper, we have approached the above problem in three steps; 1) Applied Linked data and semantic web to build Poly-Omics data as knowledge graphs and termed them as cancer decision network; 2) Reduced the dimensionality of data using Graph Pattern Mining and explained gene rewiring in cancer decision network by first time using Kirchhoff’s law for knowledge or any graph traversal; 3) Established ruled based modeling to understand the essential -Omics data from poly-Omics for breast cancer progression 4) Predicted the disease’s metastatic site using Kirchhoff’s knowledge graphs as a hidden layer in the graph convolution neural network(GCNN). The features (genes) extracted by applying Kirchhoff’s law on knowledge graphs are used to predict disease relapse site with 91.9% AUC (Area Under Curve) and performed detailed evaluation against the state-of-the-art approaches. The novelty of our approach is in the creation of RDF knowledge graphs from the poly-omics, such as the drug, disease, target(gene/protein), pathways and application of Kirchhoff’s law on knowledge graph to and the first approach to predict metastatic site from the primary tumor. Further, we have applied the rule-based knowledge graph using graph convolution neural network for metastasis site prediction makes the even classification novel.

Electronic Health Records (EHRs) contain an increasing wealth of medical information. When combined with molecular level data, they enhance the understanding of the underlying biological mechanisms of diseases, enabling the identification of key prognostic biomarkers to disease and treatment outcomes. However, the European healthcare information space is fragmented due to the lack of legal and technical standards, cost effective platforms, and sustainable business models. There is a clear need for a framework facilitating the efficient and homogenized access to anonymized distributed EHRs, merged from multiple data sources into a single data analysis space. In this paper we present the outcomes of Linked2Safety, a project that proposes a solution to these problems by providing a semantically interconnected approach to sharing aggregate data in the form of data cubes. This approach eliminates the risks associated with sharing pseudoanonymized (and therefore still personal) data while enabling the multi-source, multi-type analysis of health data through a single web based secure access platform. The Linked2Safety system is evaluated by external to the project Medical science analysts, Analytic methodology engineers and Data providers with respect to five specific dimensions of the system (analysis space, linked data space, usability of the system, legal and ethical issues, and value of the system) in this paper. For all five dimensions that were examined, the participantsâ perceptions were overwhelmingly positive. The research leading to these results was conducted as part of the project A next-generation secure linked data medical information space for semantically-interconnecting electronic health records and clinical trials systems advancing patients safety in clinical research (Linked2Safety) that received funding from the European Union’s Seventh Framework Programme (FP7/2007–2013) under Grant Agreement No 288328. peer-reviewed

Visualization of Gene Expression (GE) is a challenging task since the number of genes and their associations are difficult to predict in various set of biological studies. GE could be used to understand tissue-gene-protein relationships. Currently, Heatmaps is the standard visualization technique to depict GE data. However, Heatmaps only covers the cluster of highly dense regions. It does not provide the Interaction, Functional Annotation and pooled understanding from higher to lower expression. In the present paper, we propose a graph-based technique - based on color encoding from higher to lower expression map, along with the functional annotation. This visualization technique is highly interactive (HeatMaps are mainly static maps). The visualization system here explains the association between overlapping genes with and without tissues types. Traditional visualization techniques (viz-Heatmaps) generally explain each of the association in distinct maps. For example, overlapping genes and their interactions, based on co-expression and expression cut off are three distinct Heatmaps. We demonstrate the usability using ortholog study of GE and visualize GE using GExpressionMap. We further compare and benchmark our approach with the existing visualization techniques. It also reduces the task to cluster the expressed gene networks further to understand the over/under expression. Further, it provides the interaction based on co-expression network which itself creates co-expression clusters. GExpressionMap provides a unique graph-based visualization for GE data with their functional annotation and associated interaction among the DEGs (Differentially Expressed Genes). This publication has emanated from research conducted with the financial support of Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289, co-funded by the European Regional Development Fund. peer-reviewed 2019-12-30

A mishap in anti-cancer drug distribution is critical in breast cancer patients due to poor prediction model to identify the treatment regime in ER+ve and ER-ve (Estrogen Receptor (ER)) patients. The traditional method for the prediction depends on the change in expression across the normal-disease pair. However, it certainly misses the multidimensional aspect and underlying cause of relapse, such as various mutations, drug dosage side effects, methylation, etc. In this paper, we have developed a multi-layer neural network model to classify multidimensional genomics data into their similar annotation group. Further, we used this multi-layer cancer genomics perceptron for annotating differentially expressed genes (DEGs) to predict relapse based on ER status in breast cancer. This approach provides multivariate identification of genes, not just by differential expression, but, cause-effect of disease status due to drug overdosage and genomics-driven drug balancing method. The multi-layered neural network model, where each layer defines the relationship of similar databases with multidimensional knowledge. We illustrate that the use of multilayer knowledge graph with gene expression data for training the deep convolution neural network stratify the patient relapse and drug dosage along with underlying molecular properties. This publication has emanated from research conducted with the financial support of Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289, co-funded by the European Regional Development Fund non-peer-reviewed