Many high-performance DTA deep learning models have been proposed, but they are mostly black-box and thus lack human interpretability. Explainable AI (XAI) can make DTA models more trustworthy, and allows to distill biological knowledge from the models. Counterfactual explanation is one popular approach to explaining the behaviour of a deep neural network, which works by systematically answering the question "How would the model output change if the inputs were changed in this way?". We propose a multi-agent reinforcement learning framework, Multi-Agent Counterfactual Drug-target binding Affinity (MACDA), to generate counterfactual explanations for the drug-protein complex. Our proposed framework provides human-interpretable counterfactual instances while optimizing both the input drug and target for counterfactual generation at the same time. We benchmark the proposed MACDA framework using the Davis and PDBBind dataset and find that our framework produces more parsimonious explanations with no loss in explanation validity, as measured by encoding similarity. We then present a case study involving ABL1 and Nilotinib to demonstrate how MACDA can explain the behaviour of a DTA model in the underlying substructure interaction between inputs in its prediction, revealing mechanisms that align with prior domain knowledge.

Data augmentation plays a key role in modern machine learning pipelines. While numerous augmentation strategies have been studied in the context of computer vision and natural language processing, less is known for other data modalities. Our work extends the success of data augmentation to compositional data, i.e., simplex-valued data, which is of particular interest in microbiology, geochemistry, and other applications. Drawing on key principles from compositional data analysis, such as the \emph{Aitchison geometry of the simplex} and subcompositions, we define novel augmentation strategies for this data modality. Incorporating our data augmentations into standard supervised learning pipelines results in consistent performance gains across a wide range of standard benchmark datasets. In particular, we set a new state-of-the-art for key disease prediction tasks including colorectal cancer, type 2 diabetes, and Crohn's disease. In addition, our data augmentations enable us to define a novel contrastive learning model, which improves on previous representation learning approaches for microbiome compositional data.

MOTIVATION: The automatic discovery of sparse biomarkers that are associated with an outcome of interest is a central goal of bioinformatics. In the context of high-throughput sequencing (HTS) data, and compositional data (CoDa) more generally, an important class of biomarkers are the log-ratios between the input variables. However, identifying predictive log-ratio biomarkers from HTS data is a combinatorial optimization problem, which is computationally challenging. Existing methods are slow to run and scale poorly with the dimension of the input, which has limited their application to low- and moderate-dimensional metagenomic datasets. RESULTS: Building on recent advances from the field of deep learning, we present CoDaCoRe, a novel learning algorithm that identifies sparse, interpretable and predictive log-ratio biomarkers. Our algorithm exploits a continuous relaxation to approximate the underlying combinatorial optimization problem. This relaxation can then be optimized efficiently using the modern ML toolbox, in particular, gradient descent. As a result, CoDaCoRe runs several orders of magnitude faster than competing methods, all while achieving state-of-the-art performance in terms of predictive accuracy and sparsity. We verify the outperformance of CoDaCoRe across a wide range of microbiome, metabolite and microRNA benchmark datasets, as well as a particularly high-dimensional dataset that is outright computationally intractable for existing sparse log-ratio selection methods. AVAILABILITY AND IMPLEMENTATION: The CoDaCoRe package is available at https://github.com/egr95/R-codacore. Code and instructions for reproducing our results are available at https://github.com/cunningham-lab/codacore. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: The development of new drugs is costly, time consuming and often accompanied with safety issues. Drug repurposing can avoid the expensive and lengthy process of drug development by finding new uses for already approved drugs. In order to repurpose drugs effectively, it is useful to know which proteins are targeted by which drugs. Computational models that estimate the interaction strength of new drug-target pairs have the potential to expedite drug repurposing. Several models have been proposed for this task. However, these models represent the drugs as strings, which is not a natural way to represent molecules. We propose a new model called GraphDTA that represents drugs as graphs and uses graph neural networks to predict drug-target affinity. We show that graph neural networks not only predict drug-target affinity better than non-deep learning models, but also outperform competing deep learning methods. Our results confirm that deep learning models are appropriate for drug-target binding affinity prediction, and that representing drugs as graphs can lead to further improvements. AVAILABILITY OF IMPLEMENTATION: The proposed models are implemented in Python. Related data, pre-trained models and source code are publicly available at https://github.com/thinng/GraphDTA. All scripts and data needed to reproduce the post hoc statistical analysis are available from https://doi.org/10.5281/zenodo.3603523. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Artificial intelligence (AI) is increasingly of tremendous interest in the medical field. How-ever, failures of medical AI could have serious consequences for both clinical outcomes and the patient experience. These consequences could erode public trust in AI, which could in turn undermine trust in our healthcare institutions. This article makes 2 contributions. First, it describes the major conceptual, technical, and humanistic challenges in medical AI. Second, it proposes a solution that hinges on the education and accreditation of new expert groups who specialize in the development, verification, and operation of medical AI technologies. These groups will be required to maintain trust in our healthcare institutions.

The classification of clinical samples based on gene expression data is an important part of precision medicine. In this manuscript, we show how transforming gene expression data into a set of personalized (sample-specific) networks can allow us to harness existing graph-based methods to improve classifier performance. Existing approaches to personalized gene networks have the limitation that they depend on other samples in the data and must get re-computed whenever a new sample is introduced. Here, we propose a novel method, called Personalized Annotation-based Networks (PAN), that avoids this limitation by using curated annotation databases to transform gene expression data into a graph. Unlike competing methods, PANs are calculated for each sample independent of the population, making it a more efficient way to obtain single-sample networks. Using three breast cancer datasets as a case study, we show that PAN classifiers not only predict cancer relapse better than gene features alone, but also outperform PPI (protein-protein interactions) and population-level graph-based classifiers. This work demonstrates the practical advantages of graph-based classification for high-dimensional genomic data, while offering a new approach to making sample-specific networks. Supplementary information: PAN and the baselines are implemented in Python. Source code and data are available at https://github.com/thinng/PAN.

Abstract Interpretability allows the domain-expert to directly evaluate the model’s relevance and reliability, a practice that offers assurance and builds trust. In the healthcare setting, interpretable models should implicate relevant biological mechanisms independent of technical factors like data pre-processing. We define personalized interpretability as a measure of sample-specific feature attribution, and view it as a minimum requirement for a precision health model to justify its conclusions. Some health data, especially those generated by high-throughput sequencing experiments, have nuances that compromise precision health models and their interpretation. These data are compositional, meaning that each feature is conditionally dependent on all other features. We propose the Deep Compositional Data Analysis (DeepCoDA) framework to extend precision health modelling to high-dimensional compositional data, and to provide personalized interpretability through patient-specific weights. Our architecture maintains state-of-the-art performance across 25 real-world data sets, all while producing interpretations that are both personalized and fully coherent for compositional data.

Motivation: Although seldom acknowledged explicitly, count data generated by sequencing platforms exist as compositions for which the abundance of each component (e.g. gene or transcript) is only coherently interpretable relative to other components within that sample. This property arises from the assay technology itself, whereby the number of counts recorded for each sample is constrained by an arbitrary total sum (i.e. library size). Consequently, sequencing data, as compositional data, exist in a non-Euclidean space that, without normalization or transformation, renders invalid many conventional analyses, including distance measures, correlation coefficients and multivariate statistical models. Results: The purpose of this review is to summarize the principles of compositional data analysis (CoDA), provide evidence for why sequencing data are compositional, discuss compositionally valid methods available for analyzing sequencing data, and highlight future directions with regard to this field of study. Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: Count data generated by next-generation sequencing assays do not measure absolute transcript abundances. Instead, the data are constrained to an arbitrary "library size" by the sequencing depth of the assay, and typically must be normalized prior to statistical analysis. The constrained nature of these data means one could alternatively use a log-ratio transformation in lieu of normalization, as often done when testing for differential abundance (DA) of operational taxonomic units (OTUs) in 16S rRNA data. Therefore, we benchmark how well the ALDEx2 package, a transformation-based DA tool, detects differential expression in high-throughput RNA-sequencing data (RNA-Seq), compared to conventional RNA-Seq methods such as edgeR and DESeq2. RESULTS: To evaluate the performance of log-ratio transformation-based tools, we apply the ALDEx2 package to two simulated, and two real, RNA-Seq data sets. One of the latter was previously used to benchmark dozens of conventional RNA-Seq differential expression methods, enabling us to directly compare transformation-based approaches. We show that ALDEx2, widely used in meta-genomics research, identifies differentially expressed genes (and transcripts) from RNA-Seq data with high precision and, given sufficient sample sizes, high recall too (regardless of the alignment and quantification procedure used). Although we show that the choice in log-ratio transformation can affect performance, ALDEx2 has high precision (i.e., few false positives) across all transformations. Finally, we present a novel, iterative log-ratio transformation (now implemented in ALDEx2) that further improves performance in simulations. CONCLUSIONS: Our results suggest that log-ratio transformation-based methods can work to measure differential expression from RNA-Seq data, provided that certain assumptions are met. Moreover, these methods have very high precision (i.e., few false positives) in simulations and perform well on real data too. With previously demonstrated applicability to 16S rRNA data, ALDEx2 can thus serve as a single tool for data from multiple sequencing modalities.