Deep learning and machine learning models have shown promise in drug response prediction (DRP), yet their ability to generalize across datasets remains an open question, raising concerns about their real-world applicability. Due to the lack of standardized benchmarking approaches, model evaluations and comparisons often rely on inconsistent datasets and evaluation criteria, making it difficult to assess true predictive capabilities. In this work, we introduce a benchmarking framework for evaluating cross-dataset prediction generalization in DRP models. Our framework incorporates five publicly available drug screening datasets, seven standardized DRP models, and a scalable workflow for systematic evaluation. To assess model generalization, we introduce a set of evaluation metrics that quantify both absolute performance (e.g. predictive accuracy across datasets) and relative performance (e.g. performance drop compared to within-dataset results), enabling a more comprehensive assessment of model transferability. Our results reveal substantial performance drops when models are tested on unseen datasets, underscoring the importance of rigorous generalization assessments. While several models demonstrate relatively strong cross-dataset generalization, no single model consistently outperforms across all datasets. Furthermore, we identify CTRPv2 as the most effective source dataset for training, yielding higher generalization scores across target datasets. By sharing this standardized evaluation framework with the community, our study aims to establish a rigorous foundation for model comparison, and accelerate the development of robust DRP models for real-world applications.

Drug response prediction (DRP) methods tackle the complex task of associating the effectiveness of small molecules with the specific genetic makeup of the patient. Anti-cancer DRP is a particularly challenging task requiring costly experiments as underlying pathogenic mechanisms are broad and associated with multiple genomic pathways. The scientific community has exerted significant efforts to generate public drug screening datasets, giving a path to various machine learning models that attempt to reason over complex data space of small compounds and biological characteristics of tumors. However, the data depth is still lacking compared to application domains like computer vision or natural language processing domains, limiting current learning capabilities. To combat this issue and improves the generalizability of the DRP models, we are exploring strategies that explicitly address the imbalance in the DRP datasets. We reframe the problem as a multi-objective optimization across multiple drugs to maximize deep learning model performance. We implement this approach by constructing Multi-Objective Optimization Regularized by Loss Entropy loss function and plugging it into a Deep Learning model. We demonstrate the utility of proposed drug discovery methods and make suggestions for further potential application of the work to achieve desirable outcomes in the healthcare field.

Machine learning has made significant advancements in precision medicine, resulting in the development of various deep learning applications. For instance, in cancer drug response prediction, numerous deep learning models have been created. However, comparing these models across vast configurations of hyperparameters and data sets can be challenging. In this paper, we introduce a new scalable workflow suite that aims to answer questions that arise when comparing different models developed by different teams on similar or the same problems. We explain the problem in more detail and discuss our approach using near-exascale or exascale computers.

To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross-validation within a single study to assess model accuracy. While an essential first step, cross-validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics of Drug Sensitivity in Cancer, Cancer Cell Line Encyclopedia and Genentech Cell Line Screening Initiative (gCSI). Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.

The drug discovery process currently employed in the pharmaceutical industry typically requires about 10 years and $2–3 billion to deliver one new drug. This is both too expensive and too slow, especially in emergencies like the COVID-19 pandemic. In silico methodologies need to be improved both to select better lead compounds, so as to improve the efficiency of later stages in the drug discovery protocol, and to identify those lead compounds more quickly. No known methodological approach can deliver this combination of higher quality and speed. Here, we describe an Integrated Modeling PipEline for COVID Cure by Assessing Better LEads (IMPECCABLE) that employs multiple methodological innovations to overcome this fundamental limitation. We also describe the computational framework that we have developed to support these innovations at scale, and characterize the performance of this framework in terms of throughput, peak performance, and scientific results. We show that individual workflow components deliver 100 × to 1000 × improvement over traditional methods, and that the integration of methods, supported by scalable infrastructure, speeds up drug discovery by orders of magnitudes. IMPECCABLE has screened ∼ 1011 ligands and has been used to discover a promising drug candidate. These capabilities have been used by the US DOE National Virtual Biotechnology Laboratory and the EU Centre of Excellence in Computational Biomedicine.

Antimicrobial resistance (AMR) is a major global health threat that affects millions of people each year. Funding agencies worldwide and the global research community have expended considerable capital and effort tracking the evolution and spread of AMR by isolating and sequencing bacterial strains and performing antimicrobial susceptibility testing (AST). For the last several years, we have been capturing these efforts by curating data from the literature and data resources and building a set of assembled bacterial genome sequences that are paired with laboratory-derived AST data. This collection currently contains AST data for over 67 000 genomes encompassing approximately 40 genera and over 100 species. In this paper, we describe the characteristics of this collection, highlighting areas where sampling is comparatively deep or shallow, and showing areas where attention is needed from the research community to improve sampling and tracking efforts. In addition to using the data to track the evolution and spread of AMR, it also serves as a useful starting point for building machine learning models for predicting AMR phenotypes. We demonstrate this by describing two machine learning models that are built from the entire dataset to show where the predictive power is comparatively high or low. This AMR metadata collection is freely available and maintained on the Bacterial and Viral Bioinformatics Center (BV-BRC) FTP site ftp://ftp.bvbrc.org/RELEASE_NOTES/PATRIC_genomes_AMR.txt.

BACKGROUND: Motivated by the size and availability of cell line drug sensitivity data, researchers have been developing machine learning (ML) models for predicting drug response to advance cancer treatment. As drug sensitivity studies continue generating drug response data, a common question is whether the generalization performance of existing prediction models can be further improved with more training data. METHODS: We utilize empirical learning curves for evaluating and comparing the data scaling properties of two neural networks (NNs) and two gradient boosting decision tree (GBDT) models trained on four cell line drug screening datasets. The learning curves are accurately fitted to a power law model, providing a framework for assessing the data scaling behavior of these models. RESULTS: The curves demonstrate that no single model dominates in terms of prediction performance across all datasets and training sizes, thus suggesting that the actual shape of these curves depends on the unique pair of an ML model and a dataset. The multi-input NN (mNN), in which gene expressions of cancer cells and molecular drug descriptors are input into separate subnetworks, outperforms a single-input NN (sNN), where the cell and drug features are concatenated for the input layer. In contrast, a GBDT with hyperparameter tuning exhibits superior performance as compared with both NNs at the lower range of training set sizes for two of the tested datasets, whereas the mNN consistently performs better at the higher range of training sizes. Moreover, the trajectory of the curves suggests that increasing the sample size is expected to further improve prediction scores of both NNs. These observations demonstrate the benefit of using learning curves to evaluate prediction models, providing a broader perspective on the overall data scaling characteristics. CONCLUSIONS: A fitted power law learning curve provides a forward-looking metric for analyzing prediction performance and can serve as a co-design tool to guide experimental biologists and computational scientists in the design of future experiments in prospective research studies.

Training scientific deep learning models requires the significant compute power of high-performance computing systems. In this paper, we analyze the performance characteristics of the benchmarks from the exploratory research project CANDLE (Cancer Distributed Learning Environment) with a focus on the hyperparameters epochs, batch sizes, and learning rates. We present the parallel methodology that uses the distributed deep learning framework Horovod to parallelize the CANDLE benchmarks. We then use scaling strategies for both epochs and batch size with linear learning rate scaling to investigate how they impact the execution time and accuracy as well as the power, energy, and scalability of the parallel CANDLE benchmarks under conditions of strong scaling and weak scaling on the IBM Power9 heterogeneous system Summit at Oak Ridge National Laboratory and the Cray XC40 Theta at Argonne National Laboratory. This study provides insights into how to set the proper numbers of epochs, batch sizes, and compute resources for these benchmarks to preserve the high accuracy and to reduce the execution time of the benchmarks. We identify the data-loading performance bottleneck and then improve the performance and energy for better scalability. Results with the modified benchmarks on Summit indicate up to 78.25% in performance improvement and up to 78% in energy saving under strong scaling on up to 384 GPUs, and up to 79.5% in performance improvement and up to 77.11% in energy saving under weak scaling on up to 3,072 GPUs. On Theta, we achieve up to 45.22% performance improvement and up to 41.78% in energy saving under strong scaling on up to 384 nodes. Moreover, the modification dramatically reduces the broadcast overhead.

Cancer is a complex disease, the understanding and treatment of which are being aided through increases in the volume of collected data and in the scale of deployed computing power. Consequently, there is a growing need for the development of data-driven and, in particular, deep learning methods for various tasks such as cancer diagnosis, detection, prognosis, and prediction. Despite recent successes, however, designing high-performing deep learning models for nonimage and nontext cancer data is a time-consuming, trial-and-error, manual task that requires both cancer domain and deep learning expertise. To that end, we develop a reinforcement-learning-based neural architecture search to automate deep-learning-based predictive model development for a class of representative cancer data. We develop custom building blocks that allow domain experts to incorporate the cancer-data-specific characteristics. We show that our approach discovers deep neural network architectures that have significantly fewer trainable parameters, shorter training time, and accuracy similar to or higher than those of manually designed architectures. We study and demonstrate the scalability of our approach on up to 1,024 Intel Knights Landing nodes of the Theta supercomputer at the Argonne Leadership Computing Facility.

The Pathosystems Resource Integration Center (PATRIC, www.patricbrc.org) is designed to provide researchers with the tools and services that they need to perform genomic and other 'omic' data analyses. In response to mounting concern over antimicrobial resistance (AMR), the PATRIC team has been developing new tools that help researchers understand AMR and its genetic determinants. To support comparative analyses, we have added AMR phenotype data to over 15 000 genomes in the PATRIC database, often assembling genomes from reads in public archives and collecting their associated AMR panel data from the literature to augment the collection. We have also been using this collection of AMR metadata to build machine learning-based classifiers that can predict the AMR phenotypes and the genomic regions associated with resistance for genomes being submitted to the annotation service. Likewise, we have undertaken a large AMR protein annotation effort by manually curating data from the literature and public repositories. This collection of 7370 AMR reference proteins, which contains many protein annotations (functional roles) that are unique to PATRIC and RAST, has been manually curated so that it projects stably across genomes. The collection currently projects to 1 610 744 proteins in the PATRIC database. Finally, the PATRIC Web site has been expanded to enable AMR-based custom page views so that researchers can easily explore AMR data and design experiments based on whole genomes or individual genes.

BACKGROUND: Current multi-petaflop supercomputers are powerful systems, but present challenges when faced with problems requiring large machine learning workflows. Complex algorithms running at system scale, often with different patterns that require disparate software packages and complex data flows cause difficulties in assembling and managing large experiments on these machines. RESULTS: This paper presents a workflow system that makes progress on scaling machine learning ensembles, specifically in this first release, ensembles of deep neural networks that address problems in cancer research across the atomistic, molecular and population scales. The initial release of the application framework that we call CANDLE/Supervisor addresses the problem of hyper-parameter exploration of deep neural networks. CONCLUSIONS: Initial results demonstrating CANDLE on DOE systems at ORNL, ANL and NERSC (Titan, Theta and Cori, respectively) demonstrate both scaling and multi-platform execution.

BACKGROUND: The National Cancer Institute drug pair screening effort against 60 well-characterized human tumor cell lines (NCI-60) presents an unprecedented resource for modeling combinational drug activity. RESULTS: We present a computational model for predicting cell line response to a subset of drug pairs in the NCI-ALMANAC database. Based on residual neural networks for encoding features as well as predicting tumor growth, our model explains 94% of the response variance. While our best result is achieved with a combination of molecular feature types (gene expression, microRNA and proteome), we show that most of the predictive power comes from drug descriptors. To further demonstrate value in detecting anticancer therapy, we rank the drug pairs for each cell line based on model predicted combination effect and recover 80% of the top pairs with enhanced activity. CONCLUSIONS: We present promising results in applying deep learning to predicting combinational drug response. Our feature analysis indicates screening data involving more cell lines are needed for the models to make better use of molecular features.

Metagenomic sequencing has produced significant amounts of data in recent years. For example, as of summer 2013, MG-RAST has been used to annotate over 110,000 data sets totaling over 43 Terabases. With metagenomic sequencing finding even wider adoption in the scientific community, the existing web-based analysis tools and infrastructure in MG-RAST provide limited capability for data retrieval and analysis, such as comparative analysis between multiple data sets. Moreover, although the system provides many analysis tools, it is not comprehensive. By opening MG-RAST up via a web services API (application programmers interface) we have greatly expanded access to MG-RAST data, as well as provided a mechanism for the use of third-party analysis tools with MG-RAST data. This RESTful API makes all data and data objects created by the MG-RAST pipeline accessible as JSON objects. As part of the DOE Systems Biology Knowledgebase project (KBase, http://kbase.us) we have implemented a web services API for MG-RAST. This API complements the existing MG-RAST web interface and constitutes the basis of KBase's microbial community capabilities. In addition, the API exposes a comprehensive collection of data to programmers. This API, which uses a RESTful (Representational State Transfer) implementation, is compatible with most programming environments and should be easy to use for end users and third parties. It provides comprehensive access to sequence data, quality control results, annotations, and many other data types. Where feasible, we have used standards to expose data and metadata. Code examples are provided in a number of languages both to show the versatility of the API and to provide a starting point for users. We present an API that exposes the data in MG-RAST for consumption by our users, greatly enhancing the utility of the MG-RAST service.

BACKGROUND: Biology is rapidly becoming a data intensive, data-driven science. It is essential that data is represented and connected in ways that best represent its full conceptual content and allows both automated integration and data driven decision-making. Recent advancements in distributed multi-relational directed graphs, implemented in the form of the Semantic Web make it possible to deal with complicated heterogeneous data in new and interesting ways. RESULTS: This paper presents a new approach, scenario driven data modelling (SDDM), that integrates multi-relational directed graphs with data streams. SDDM can be applied to virtually any data integration challenge with widely divergent types of data and data streams. In this work, we explored integrating genetics data with reports from traditional media. SDDM was applied to the New Delhi metallo-beta-lactamase gene (NDM-1), an emerging global health threat. The SDDM process constructed a scenario, created a RDF multi-relational directed graph that linked diverse types of data to the Semantic Web, implemented RDF conversion tools (RDFizers) to bring content into the Sematic Web, identified data streams and analytical routines to analyse those streams, and identified user requirements and graph traversals to meet end-user requirements. CONCLUSIONS: We provided an example where SDDM was applied to a complex data integration challenge. The process created a model of the emerging NDM-1 health threat, identified and filled gaps in that model, and constructed reliable software that monitored data streams based on the scenario derived multi-relational directed graph. The SDDM process significantly reduced the software requirements phase by letting the scenario and resulting multi-relational directed graph define what is possible and then set the scope of the user requirements. Approaches like SDDM will be critical to the future of data intensive, data-driven science because they automate the process of converting massive data streams into usable knowledge.

BACKGROUND: Traditional genome annotation systems were developed in a very different computing era, one where the World Wide Web was just emerging. Consequently, these systems are built as centralized black boxes focused on generating high quality annotation submissions to GenBank/EMBL supported by expert manual curation. The exponential growth of sequence data drives a growing need for increasingly higher quality and automatically generated annotation. Typical annotation pipelines utilize traditional database technologies, clustered computing resources, Perl, C, and UNIX file systems to process raw sequence data, identify genes, and predict and categorize gene function. These technologies tightly couple the annotation software system to hardware and third party software (e.g. relational database systems and schemas). This makes annotation systems hard to reproduce, inflexible to modification over time, difficult to assess, difficult to partition across multiple geographic sites, and difficult to understand for those who are not domain experts. These systems are not readily open to scrutiny and therefore not scientifically tractable. The advent of Semantic Web standards such as Resource Description Framework (RDF) and OWL Web Ontology Language (OWL) enables us to construct systems that address these challenges in a new comprehensive way. RESULTS: Here, we develop a framework for linking traditional data to OWL-based ontologies in genome annotation. We show how data standards can decouple hardware and third party software tools from annotation pipelines, thereby making annotation pipelines easier to reproduce and assess. An illustrative example shows how TURTLE (Terse RDF Triple Language) can be used as a human readable, but also semantically-aware, equivalent to GenBank/EMBL files. CONCLUSIONS: The power of this approach lies in its ability to assemble annotation data from multiple databases across multiple locations into a representation that is understandable to researchers. In this way, all researchers, experimental and computational, will more easily understand the informatics processes constructing genome annotation and ultimately be able to help improve the systems that produce them.

Abstract Summary: We have developed two novel methods for Singular Value Decomposition analysis (SVD) of microarray data. The first is a threshold-based method for obtaining gene groups, and the second is a method for obtaining a measure of confidence in SVD analysis. Gene groups are obtained by identifying elements of the left singular vectors, or gene coefficient vectors, that are greater in magnitude than the threshold \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(WN^{{-}1/2}\) \end{document}, where \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(N\) \end{document}is the number of genes, and \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(W\) \end{document}is a weight factor whose default value is 3. The groups are non-exclusive and may contain genes of opposite (i.e. inversely correlated) regulatory response. The confidence measure is obtained by systematically deleting assays from the data set, interpolating the SVD of the reduced data set to reconstruct the missing assay, and calculating the Pearson correlation between the reconstructed assay and the original data. This confidence measure is applicable when each experimental assay corresponds to a value of parameter that can be interpolated, such as time, dose or concentration. Algorithms for the grouping method and the confidence measure are available in a software application called SVD Microarray ANalysis (SVDMAN). In addition to calculating the SVD for generic analysis, SVDMAN provides a new means for using microarray data to develop hypotheses for gene associations and provides a measure of confidence in the hypotheses, thus extending current SVD research in the area of global gene expression analysis. Availability: ftp://bpublic.lanl.gov/compbio/software Contact: [email protected] Supplementary information: http://home.lanl.gov/svdman * To whom correspondence should be addressed.