• Online preprint

Abstract

In constrained real-world scenarios where it is challenging or costly to generate data, disciplined methods for acquiring informative new data points are of fundamental importance for the efficient training of machine learning (ML) models. Active learning (AL) is a subfield of ML focused on the development of methods to iteratively and economically acquire data through strategically querying new data points that are the most useful for a particular task. Here, we introduce PyRelationAL, an open source library for AL research. We describe a modular toolkit that is compatible with diverse ML frameworks (e.g. PyTorch, Scikit-Learn, TensorFlow, JAX). Furthermore, to help accelerate research and development in the field, the library implements a number of published methods and provides API access to wide-ranging benchmark datasets and AL task configurations based on existing literature. The library is supplemented by an expansive set of tutorials, demos, and documentation to help users get started. We perform experiments on the PyRelationAL collection of benchmark datasets and showcase the considerable economies that AL can provide. PyRelationAL is maintained using modern software engineering practices - with an inclusive contributor code of conduct - to promote long term library quality and utilisation.

• Online preprint

Abstract

Selecting optimal drug repurposing combinations for further preclinical development is a challenging technical feat. Due to the toxicity of many therapeutic agents (e.g., chemotherapy), practitioners have favoured selection of synergistic compounds whereby lower doses can be used whilst maintaining high efficacy. For a fixed small molecule library, an exhaustive combinatorial chemical screen becomes infeasible to perform for academic and industry laboratories alike. Deep learning models have achieved state-of-the-art results in silico for the prediction of synergy scores. However, databases of drug combinations are highly biased towards synergistic agents and these results do not necessarily generalise out of distribution. We employ a sequential model optimization search applied to a deep learning model to quickly discover highly synergistic drug combinations active against a cancer cell line, while requiring substantially less screening than an exhaustive evaluation. Through iteratively adapting the model to newly acquired data, after only 3 rounds of ML-guided experimentation (including a calibration round), we find that the set of combinations queried by our model is enriched for highly synergistic combinations. Remarkably, we rediscovered a synergistic drug combination that was later confirmed to be under study within clinical trials.

• Online version

Abstract

Graph Machine Learning (GML) is receiving growing interest within the pharmaceutical and biotechnology industries for its ability to model biomolecular structures, the functional relationships between them, and integrate multi-omic datasets-amongst other data types. Herein, we present a multidisciplinary academic-industrial review of the topic within the context of drug discovery and development. After introducing key terms and modelling approaches, we move chronologically through the drug development pipeline to identify and summarise work incorporating: target identification, design of small molecules and biologics, and drug repurposing. Whilst the field is still emerging, key milestones including repurposed drugs entering in vivo studies, suggest graph machine learning will become a modelling framework of choice within biomedical machine learning.

• Online version

Abstract

With the advancement of high-throughput biotechnologies, we increasingly accumulate biomedical data about diseases, especially cancer. There is a need for computational models and methods to sift through, integrate, and extract new knowledge from the diverse available data to improve the mechanistic understanding of diseases and patient care. To uncover molecular mechanisms and drug indications for specific cancer types, we develop an integrative framework able to harness a wide range of diverse molecular and pan-cancer data. We show that our approach outperforms competing methods and can identify new associations. Furthermore, through the joint integration of data sources, our framework can also uncover links between cancer types and molecular entities for which no prior knowledge is available. Our new framework is flexible and can be easily reformulated to study any biomedical problems.

• Online version

Abstract

Biological and cellular systems are often modeled as graphs in which vertices represent objects of interest (genes, proteins, drugs) and edges represent relational ties between these objects (binds-to, interacts-with, regulates). This approach has been highly successful owing to the theory, methodology and software that support analysis and learning on graphs. Graphs, however, suffer from information loss when modeling physical systems due to their inability to accurately represent multi-object relationships. Hypergraphs, a generalization of graphs, provide a framework to mitigate information loss and unify disparate graph-based methodologies. We present a hypergraph-based approach for modeling biological systems and formulate vertex classification, edge classification and link prediction problems on (hyper)graphs as instances of vertex classification on (extended, dual) hypergraphs. We then introduce a novel kernel method on vertex- and edge-labeled (colored) hypergraphs for analysis and learning. The method is based on exact and inexact (via hypergraph edit distances) enumeration of hypergraphlets; i.e., small hypergraphs rooted at a vertex of interest. We empirically evaluate this method on fifteen biological networks and show its potential use in a positive-unlabeled setting to estimate the interactome sizes in various species.

• Online version

Abstract

Diseases involve complex modifications to the cellular machinery. The gene expression profile of the affected cells contains characteristic patterns linked to a disease. Hence, new biological knowledge about a disease can be extracted from these profiles, improving our ability to diagnose and assess disease risks. This knowledge can be used for drug re-purposing, or by physicians to evaluate a patient's condition and co-morbidity risk.

Here, we consider differential gene expressions obtained by microarray technology for patients diagnosed with various diseases. Based on these data and cellular multi-scale organization, we aim at uncovering disease--disease, disease--gene and disease--pathway associations. We propose a neural network with structure based on the multi-scale organization of proteins in a cell into biological pathways. We show that this model is able to correctly predict the diagnosis for the majority of patients. Through the analysis of the trained model, we predict disease--disease, disease--pathway, and disease--gene associations and validate the predictions by comparisons to known interactions and literature search, proposing putative explanations for the predictions.

• Ebook link

Abstract

Hardly any entity exists in absolute isolation of others. It is usually a part of a larger system of greater complexity. This remains true at any scale. For instance, stars and planets interact via different forces of attraction and repulsion that shape the universe. Species interact tightly through predator-prey relationships to form an often fragile ecosystem. Neurons in a brain interact to carry out complex tasks ranging from limb motion to information processing. Studying single entities in isolation of others is crucial to our understanding. However, it is also essential to study their roles as part of wider systems of interacting components. The organisation of such systems is far from random and reflects the function of the system. Understanding the organisational principles of the systems can shed light on the impact of changes on the system or on its environment.

Graphs (networks) are means of representing entities and their interactions, while abstracting much of the data about them. These representations are needed to develop tools to investigate the organisation of complex systems. Everyday examples of network representations of real systems include maps of public transport networks, which connect various stations, and the World Wide Web. In biology, many systems can naturally be modeled by networks.

In this book chapter, we introduce elements of graph and network theory. The target audience is students enrolled in bioinformatics modules.

• Online version

Abstract

Molecular interactions have widely been modelled as networks. The local wiring patterns around molecules in molecular networks are linked with their biological functions. However, networks model only pairwise interactions between molecules and cannot explicitly and directly capture the higher-order molecular organization, such as protein complexes and pathways. Hence, we ask if hypergraphs (hypernetworks), that directly capture entire complexes and pathways along with protein–protein interactions (PPIs), carry additional functional information beyond what can be uncovered from networks of pairwise molecular interactions. The mathematical formalism of a hypergraph has long been known, but not often used in studying molecular networks due to the lack of sophisticated algorithms for mining the underlying biological information hidden in the wiring patterns of molecular systems modelled as hypernetworks.

We propose a new, multi-scale, protein interaction hypernetwork model that utilizes hypergraphs to capture different scales of protein organization, including PPIs, protein complexes and pathways. In analogy to graphlets, we introduce hypergraphlets, small, connected, non-isomorphic, induced sub-hypergraphs of a hypergraph, to quantify the local wiring patterns of these multi-scale molecular hypergraphs and to mine them for new biological information. We apply them to model the multi-scale protein networks of bakers yeast and human and show that the higher-order molecular organization captured by these hypergraphs is strongly related to the underlying biology. Importantly, we demonstrate that our new models and data mining tools reveal different, but complementary biological information compared with classical PPI networks. We apply our hypergraphlets to successfully predict biological functions of uncharacterized proteins.

• Online version

Abstract

Vapor condensation is a crucial part of a broad range of industrial applications including power generation, water harvesting, and air conditioning. Hydrophobic and superhydrophobic surfaces promote dropwise condensation in vapor-filled environments and increase their heat transfer coefficients more than filmwise condensation on hydrophilic surfaces. Although dropwise condensation can lead to energy-efficient transfer, it is hard to achieve stable dropwise condensation in high-temperature environments. To decide the best conditions for achieving higher heat transfer is also difficult because the heat transfer coefficient is influenced by not only surface wettability but also surface structures of thin films and substrates. Herein, we fabricated thin films with different wettabilities and surface structures using polytetrafluoroethylene (PTFE) which show high heat resistance to determine the best conditions for heat transfer. Several different films were prepared by electrospinning a mixed solution of PTFE and polyvinyl alcohol on aluminum (Al) and copper (Cu) tubes. After annealing them, the PTFE thin films enhanced heat transfer performance and showed stable dropwise condensation in high-temperature environments. The films fabricated by electrospinning a solution containing 66 wt% PTFE displayed the highest heat transfer coefficients, with heat transfer coefficients 64% and 61% greater than those of uncoated Al and Cu tubes, respectively. That is because homogeneous superhydrophobic surfaces that showed the highest departure frequency of condensed water droplets were fabricated using 66 wt% PTFE. The results suggest that these electrospun PTFE thin films would demonstrate excellent potential for use on the surface of heat exchangers in various industries.