Search

Scientific results should not just be ‘repeatable’ (replicable in the same laboratory under identical conditions), but also ‘reproducible’ (replicable in other laboratories under similar conditions). Results should also, if possible, be ‘robust’ (replicable under a wide range of conditions). The reproducibility and robustness of only a small fraction of published biomedical results has been tested; furthermore, when reproducibility is tested, it is often not found. This situation is termed ‘the reproducibility crisis', and it is one the most important issues facing biomedicine. This crisis would be solved if it were possible to automate reproducibility testing. Here, we describe the semi-automated testing for reproducibility and robustness of simple statements (propositions) about cancer cell biology automatically extracted from the literature. From 12 260 papers, we automatically extracted statements predicted to describe experimental results regarding a change of gene expression in response to drug treatment in breast cancer, from these we selected 74 statements of high biomedical interest. To test the reproducibility of these statements, two different teams used the laboratory automation system Eve and two breast cancer cell lines (MCF7 and MDA-MB-231). Statistically significant evidence for repeatability was found for 43 statements, and significant evidence for reproducibility/robustness in 22 statements. In two cases, the automation made serendipitous discoveries. The reproduced/robust knowledge provides significant insight into cancer. We conclude that semi-automated reproducibility testing is currently achievable, that it could be scaled up to generate a substantive source of reliable knowledge and that automation has the potential to mitigate the reproducibility crisis.

Significance Machine learning (ML) is the branch of artificial intelligence (AI) that develops computational systems that learn from experience. In supervised ML, the ML system generalizes from labelled examples to learn a model that can predict the labels of unseen examples. Examples are generally represented using features that directly describe the examples. For instance, in drug design, ML uses features that describe molecular shape and so on. In cases where there are multiple related ML problems, it is possible to use a different type of feature: predictions made about the examples by ML models learned on other problems. We call this transformational ML. We show that this results in better predictions and improved understanding when applied to scientific problems., Almost all machine learning (ML) is based on representing examples using intrinsic features. When there are multiple related ML problems (tasks), it is possible to transform these features into extrinsic features by first training ML models on other tasks and letting them each make predictions for each example of the new task, yielding a novel representation. We call this transformational ML (TML). TML is very closely related to, and synergistic with, transfer learning, multitask learning, and stacking. TML is applicable to improving any nonlinear ML method. We tested TML using the most important classes of nonlinear ML: random forests, gradient boosting machines, support vector machines, k-nearest neighbors, and neural networks. To ensure the generality and robustness of the evaluation, we utilized thousands of ML problems from three scientific domains: drug design, predicting gene expression, and ML algorithm selection. We found that TML significantly improved the predictive performance of all the ML methods in all the domains (4 to 50% average improvements) and that TML features generally outperformed intrinsic features. Use of TML also enhances scientific understanding through explainable ML. In drug design, we found that TML provided insight into drug target specificity, the relationships between drugs, and the relationships between target proteins. TML leads to an ecosystem-based approach to ML, where new tasks, examples, predictions, and so on synergistically interact to improve performance. To contribute to this ecosystem, all our data, code, and our ∼50,000 ML models have been fully annotated with metadata, linked, and openly published using Findability, Accessibility, Interoperability, and Reusability principles (∼100 Gbytes).

The ongoing COVID-19 pandemic urges searches for antiviral agents that can block infection or ameliorate its symptoms. Using dissimilar search strategies for new antivirals will improve our overall chances of finding effective treatments. Here, we have established an experimental platform for screening of small molecule inhibitors of SARS-CoV-2 main protease in Saccharomyces cerevisiae cells, genetically engineered to enhance cellular uptake of small molecules in the environment. The system consists of a fusion of the E. coli toxin MazF and its antitoxin MazE, with insertion of a protease cleavage site in the linker peptide connecting the MazE and MazF moieties. Expression of the viral protease confers cleavage of the MazEF fusion, releasing the MazF toxin from its antitoxin, resulting in growth inhibition. In the presence of a small molecule inhibiting the protease, cleavage is blocked and the MazF toxin remains inhibited, promoting growth. The system thus allows positive selection for inhibitors. The engineered yeast strain is tagged with a fluorescent marker protein, allowing precise monitoring of its growth in the presence or absence of inhibitor. We detect an established main protease inhibitor down to 10 μM by a robust growth increase. The system is suitable for robotized large-scale screens. It allows in vivo evaluation of drug candidates, and is rapidly adaptable for new variants of the protease with deviant site specificities.IMPORTANCEThe COVID-19 pandemic may continue several years before vaccination campaigns can put an end it globally. Thus, the need for discovery of new antiviral drug candidates will remain. We have engineered a system in yeast cells for detection of small molecule inhibitors of one attractive drug target of SARS-CoV-2, its main protease which is required for viral replication. To detect inhibitors in live cells brings the advantage that only compounds capable of entering the cell and remain stable there, will score in the system. Moreover, by its design in yeast, the system is rapidly adaptable for tuning of detection level, eventual modification of protease cleavage site in case of future mutant variants of the SARS-CoV-2 main protease, or even for other proteases.

In phenotype prediction the physical characteristics of an organism are predicted from knowledge of its genotype and environment. Such studies, often called genome-wide association studies, are of the highest societal importance, as they are of central importance to medicine, crop-breeding, etc. We investigated three phenotype prediction problems: one simple and clean (yeast), and the other two complex and real-world (rice and wheat). We compared standard machine learning methods; elastic net, ridge regression, lasso regression, random forest, gradient boosting machines (GBM), and support vector machines (SVM), with two state-of-the-art classical statistical genetics methods; genomic BLUP and a two-step sequential method based on linear regression. Additionally, using the clean yeast data, we investigated how performance varied with the complexity of the biological mechanism, the amount of observational noise, the number of examples, the amount of missing data, and the use of different data representations. We found that for almost all the phenotypes considered, standard machine learning methods outperformed the methods from classical statistical genetics. On the yeast problem, the most successful method was GBM, followed by lasso regression, and the two statistical genetics methods; with greater mechanistic complexity GBM was best, while in simpler cases lasso was superior. In the wheat and rice studies the best two methods were SVM and BLUP. The most robust method in the presence of noise, missing data, etc. was random forests. The classical statistical genetics method of genomic BLUP was found to perform well on problems where there was population structure. This suggests that standard machine learning methods need to be refined to include population structure information when this is present. We conclude that the application of machine learning methods to phenotype prediction problems holds great promise, but that determining which methods is likely to perform well on any given problem is elusive and non-trivial.

We developed an in-silico model of cancer growth to investigate the extent to which metabolic switching occurs in tumour masses. Cancer therapies based on glycoconjugation, the linking of a drug to glucose or another sugar, allow improved selectivity and targeting, thus reducing harmful side effects. This mechanism exploits the over-expression of glucose membrane transporters, a phenotypic alteration in cancer cells included in an array of metabolic alterations known as the Warburg effect. However, the extent to which tumour masses adopt the Warburg phenotype is unclear, potentially limiting the efficacy of therapies based on glycoconjugation. We simulated multiple “what-if” scenarios, each modelling increasing proportions of tumour populations that adopted the Warburg phenotype, and compared the results to the expected growth curves derived from laboratory studies. Our results suggest that the Warburg phenotype is prevalent in tumours, with the population of cancer cells adopting this phenotype significantly outnumbering that of cells that do not.

Machine reading (MR) is essential for unlocking valuable knowledge contained in millions of existing biomedical documents. Over the last two decades1,2, the most dramatic advances in MR have followed in the wake of critical corpus development3. Large, well-annotated corpora have been associated with punctuated advances in MR methodology and automated knowledge extraction systems in the same way that ImageNet4 was fundamental for developing machine vision techniques. This study contributes six components to an advanced, named entity analysis tool for biomedicine: (a) a new, Named Entity Recognition Ontology (NERO) developed specifically for describing textual entities in biomedical texts, which accounts for diverse levels of ambiguity, bridging the scientific sublanguages of molecular biology, genetics, biochemistry, and medicine; (b) detailed guidelines for human experts annotating hundreds of named entity classes; (c) pictographs for all named entities, to simplify the burden of annotation for curators; (d) an original, annotated corpus comprising 35,865 sentences, which encapsulate 190,679 named entities and 43,438 events connecting two or more entities; (e) validated, off-the-shelf, named entity recognition (NER) automated extraction, and; (f) embedding models that demonstrate the promise of biomedical associations embedded within this corpus.

Rationale: Pulmonary endothelial permeability contributes to the high-permeability pulmonary edema that characterizes acute respiratory distress syndrome. Circulating BMP9 (bone morphogenetic protein 9) is emerging as an important regulator of pulmonary vascular homeostasis. Objectives:To determine whether endogenous BMP9 plays a role in preserving pulmonary endothelial integrity and whether loss of endogenous BMP9 occurs during LPS challenge. Methods: A BMP9-neutralizing antibody was administrated to healthy adult mice, and lung vasculature was examined. Potential mechanisms were delineated by transcript analysis in human lung endothelial cells. The impact of BMP9 administration was evaluated in a murine acute lung injury model induced by inhaled LPS. Levels of BMP9 were measured in plasma from patients with sepsis and from endotoxemic mice. Measurements and Main Results: Subacute neutralization of endogenous BMP9 in mice (N = 12) resulted in increased lung vascular permeability (P = 0.022), interstitial edema (P = 0.0047), and neutrophil extravasation (P = 0.029) compared with IgG control treatment (N = 6). In pulmonary endothelial cells, BMP9 regulated transcriptome pathways implicated in vascular permeability and cell-membrane integrity. Augmentation of BMP9 signaling in mice (N = 8) prevented inhaled LPS-induced lung injury (P = 0.0027) and edema (P < 0.0001). In endotoxemic mice (N = 12), endogenous circulating BMP9 concentrations were markedly reduced, the causes of which include a transient reduction in hepatic BMP9 mRNA expression and increased elastase activity in plasma. In human patients with sepsis (N = 10), circulating concentratons of BMP9 were also markedly reduced (P < 0.0001). Conclusions: Endogenous circulating BMP9 is a pulmonary endothelial-protective factor, downregulated during inflammation. Exogenous BMP9 offers a potential therapy to prevent increased pulmonary endothelial permeability in lung injury.

Machine reading is essential for unlocking valuable knowledge contained in the millions of existing biomedical documents. Over the last two decades1,2, the most dramatic advances in machine-reading have followed in the wake of critical corpus development3. Large, well-annotated corpora have been associated with punctuated advances in machine reading methodology and automated knowledge extraction systems in the same way that ImageNet4was fundamental for developing machine vision techniques. This study contributes six components to an advanced, named-entity analysis tool for biomedicine: (a) a new, Named-Entity Recognition Ontology (NERO) developed specifically for describing entities in biomedical texts, which accounts for diverse levels of ambiguity, bridging the scientific sublanguages of molecular biology, genetics, biochemistry, and medicine; (b) detailed guidelines for human experts annotating hundreds of named-entity classes; (c) pictographs for all named entities, to simplify the burden of annotation for curators; (d) an original, annotated corpus comprising 35,865 sentences, which encapsulate 190,679 named entities and 43,438 events connecting two or more entities; (e) validated, off-the-shelf, named-entity recognition automated extraction, and; (f) embedding models that demonstrate the promise of biomedical associations embedded within this corpus.

This paper presents a combination of machine learning techniques for slip detection in grasping, based on temporal features collected by low-cost tactile sensors. A slippage is an event that is subsequent to prior micro-slippages that have occurred at hand-object contact. The method is based on the application of a sequential classification technique (a variant of recurrent neural networks known as long short-term memory networks or LSTMs), whereby time-series pressure readings from tactile sensors are classified as either slip or non-slip events. We also propose a novel method for autonomous labelling, removing the need for humans in the labelling process. Lastly, this paper proposes a new design for an adaptable wearable tactile sensing device that integrates non-expensive sensors. Our proposed method achieved high accuracy in the classification of slip and non-slip events, obtaining over 95% in offline classification and 89% in online classification using a Sawyer robot.