Chapter 2: Data-driven view of disease biology

Finding relationships among biological entities

Confusion over the concepts of “relationships” and “similarities” lies at the heart of many battles over the direction and intent of research projects. Here is a short story that demonstrates the difference between the two concepts: You look up at the clouds, and you begin to see the shape of a lion. The cloud has a tail, like a lion’s tale, and a fluffy head, like a lion’s mane. With a little imagination the mouth of the lion seems to roar down from the sky. You have succeeded in finding similarities between the cloud and a lion. If you look at a cloud and you imagine a tea kettle producing a head of steam and you recognize that the physical forces that create a cloud and the physical forces that produced steam from a heated kettle are the same, then you have found a relationship. Most popular classification algorithms operate by grouping together data objects that have similar properties or values. In so doing, they may miss finding the true relationships among objects. Traditionally, relationships among data objects are discovered by an intellectual process. In this chapter, we will discuss the scientific gains that come when we classify biological entities by relationships, not by their similarities.

The assessment of efficient representation of drug features using deep learning for drug repositioning

BACKGROUND

De novo drug discovery is a time-consuming and expensive process. Nowadays, drug repositioning is utilized as a common strategy to discover a new drug indication for existing drugs. This strategy is mostly used in cases with a limited number of candidate pairs of drugs and diseases. In other words, they are not scalable to a large number of drugs and diseases. Most of the in-silico methods mainly focus on linear approaches while non-linear models are still scarce for new indication predictions. Therefore, applying non-linear computational approaches can offer an opportunity to predict possible drug repositioning candidates.

RESULTS

In this study, we present a non-linear method for drug repositioning. We extract four drug features and two disease features to find the semantic relations between drugs and diseases. We utilize deep learning to extract an efficient representation for each feature. These representations reduce the dimension and heterogeneity of biological data. Then, we assess the performance of different combinations of drug features to introduce a pipeline for drug repositioning. In the available database, there are different numbers of known drug-disease associations corresponding to each combination of drug features. Our assessment shows that as the numbers of drug features increase, the numbers of available drugs decrease. Thus, the proposed method with large numbers of drug features is as accurate as small numbers.

CONCLUSION

Our pipeline predicts new indications for existing drugs systematically, in a more cost-effective way and shorter timeline. We assess the pipeline to discover the potential drug-disease associations based on cross-validation experiments and some clinical trial studies.

A data-driven interactome of synergistic genes improves network-based cancer outcome prediction

Robustly predicting outcome for cancer patients from gene expression is an important challenge on the road to better personalized treatment. Network-based outcome predictors (NOPs), which considers the cellular wiring diagram in the classification, hold much promise to improve performance, stability and interpretability of identified marker genes. Problematically, reports on the efficacy of NOPs are conflicting and for instance suggest that utilizing random networks performs on par to networks that describe biologically relevant interactions. In this paper we turn the prediction problem around: instead of using a given biological network in the NOP, we aim to identify the network of genes that truly improves outcome prediction. To this end, we propose SyNet, a gene network constructed ab initio from synergistic gene pairs derived from survival-labelled gene expression data. To obtain SyNet, we evaluate synergy for all 69 million pairwise combinations of genes resulting in a network that is specific to the dataset and phenotype under study and can be used to in a NOP model. We evaluated SyNet and 11 other networks on a compendium dataset of >4000 survival-labelled breast cancer samples. For this purpose, we used cross-study validation which more closely emulates real world application of these outcome predictors. We find that SyNet is the only network that truly improves performance, stability and interpretability in several existing NOPs. We show that SyNet overlaps significantly with existing gene networks, and can be confidently predicted (~85% AUC) from graph-topological descriptions of these networks, in particular the breast tissue-specific network. Due to its data-driven nature, SyNet is not biased to well-studied genes and thus facilitates post-hoc interpretation. We find that SyNet is highly enriched for known breast cancer genes and genes related to e.g. histological grade and tamoxifen resistance, suggestive of a role in determining breast cancer outcome.

Applications of Deep Learning in Biomedicine

Increases in throughput and installed base of biomedical research equipment led to a massive accumulation of -omics data known to be highly variable, high-dimensional, and sourced from multiple often incompatible data platforms. While this data may be useful for biomarker identification and drug discovery, the bulk of it remains underutilized. Deep neural networks (DNNs) are efficient algorithms based on the use of compositional layers of neurons, with advantages well matched to the challenges -omics data presents. While achieving state-of-the-art results and even surpassing human accuracy in many challenging tasks, the adoption of deep learning in biomedicine has been comparatively slow. Here, we discuss key features of deep learning that may give this approach an edge over other machine learning methods. We then consider limitations and review a number of applications of deep learning in biomedical studies demonstrating proof of concept and practical utility.

Pathogen genome bioinformatics

Recent advances in DNA sequencing technology have made the whole-genome sequencing of pathogens in a clinically relevant turn-around time both technically and economically feasible. The DNA sequencing of pathogens with epidemic potential offers new and exciting opportunities for high-resolution public health surveillance. This chapter outlines major methods and bioinformatics tools for pathogen genome characterization, the identification of infectious disease clusters, as well as for genomics-guided biosurveillance. Existing challenges are also considered.

Prediction of gene-phenotype associations in humans, mice, and plants using phenologs

Background

Phenotypes and diseases may be related to seemingly dissimilar phenotypes in other species by means of the orthology of underlying genes. Such “orthologous phenotypes,” or “phenologs,” are examples of deep homology, and may be used to predict additional candidate disease genes.

Results

In this work, we develop an unsupervised algorithm for ranking phenolog-based candidate disease genes through the integration of predictions from the k nearest neighbor phenologs, comparing classifiers and weighting functions by cross-validation. We also improve upon the original method by extending the theory to paralogous phenotypes. Our algorithm makes use of additional phenotype data — from chicken, zebrafish, and E. coli, as well as new datasets for C. elegans — establishing that several types of annotations may be treated as phenotypes. We demonstrate the use of our algorithm to predict novel candidate genes for human atrial fibrillation (such as HRH2, ATP4A, ATP4B, and HOPX) and epilepsy (e.g., PAX6 and NKX2-1). We suggest gene candidates for pharmacologically-induced seizures in mouse, solely based on orthologous phenotypes from E. coli. We also explore the prediction of plant gene–phenotype associations, as for the Arabidopsis response to vernalization phenotype.

Conclusions

We are able to rank gene predictions for a significant portion of the diseases in the Online Mendelian Inheritance in Man database. Additionally, our method suggests candidate genes for mammalian seizures based only on bacterial phenotypes and gene orthology. We demonstrate that phenotype information may come from diverse sources, including drug sensitivities, gene ontology biological processes, and in situ hybridization annotations. Finally, we offer testable candidates for a variety of human diseases, plant traits, and other classes of phenotypes across a wide array of species.