Testing the reproducibility and robustness of the cancer biology literature by robot

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.

NERO: a biomedical named-entity (recognition) ontology with a large, annotated corpus reveals meaningful associations through text embedding

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.

Multi-task learning with a natural metric for quantitative structure activity relationship learning

The goal of quantitative structure activity relationship (QSAR) learning is to learn a function that, given the structure of a small molecule (a potential drug), outputs the predicted activity of the compound. We employed multi-task learning (MTL) to exploit commonalities in drug targets and assays. We used datasets containing curated records about the activity of specific compounds on drug targets provided by ChEMBL. Totally, 1091 assays have been analysed. As a baseline, a single task learning approach that trains random forest to predict drug activity for each drug target individually was considered. We then carried out feature-based and instance-based MTL to predict drug activities. We introduced a natural metric of evolutionary distance between drug targets as a measure of tasks relatedness. Instance-based MTL significantly outperformed both, feature-based MTL and the base learner, on 741 drug targets out of 1091. Feature-based MTL won on 179 occasions and the base learner performed best on 171 drug targets. We conclude that MTL QSAR is improved by incorporating the evolutionary distance between targets. These results indicate that QSAR learning can be performed effectively, even if little data is available for specific drug targets, by leveraging what is known about similar drug targets.

Meta-QSAR: a large-scale application of meta-learning to drug design and discovery

We investigate the learning of quantitative structure activity relationships (QSARs) as a case-study of meta-learning. This application area is of the highest societal importance, as it is a key step in the development of new medicines. The standard QSAR learning problem is: given a target (usually a protein) and a set of chemical compounds (small molecules) with associated bioactivities (e.g. inhibition of the target), learn a predictive mapping from molecular representation to activity. Although almost every type of machine learning method has been applied to QSAR learning there is no agreed single best way of learning QSARs, and therefore the problem area is well-suited to meta-learning. We first carried out the most comprehensive ever comparison of machine learning methods for QSAR learning: 18 regression methods, 3 molecular representations, applied to more than 2700 QSAR problems. (These results have been made publicly available on OpenML and represent a valuable resource for testing novel meta-learning methods.) We then investigated the utility of algorithm selection for QSAR problems. We found that this meta-learning approach outperformed the best individual QSAR learning method (random forests using a molecular fingerprint representation) by up to 13%, on average. We conclude that meta-learning outperforms base-learning methods for QSAR learning, and as this investigation is one of the most extensive ever comparisons of base and meta-learning methods ever made, it provides evidence for the general effectiveness of meta-learning over base-learning.

Thematic issue of the Second combined Bio-ontologies and Phenotypes Workshop

This special issue covers selected papers from the 18th Bio-Ontologies Special Interest Group meeting and Phenotype Day, which took place at the Intelligent Systems for Molecular Biology (ISMB) conference in Dublin in 2015. The papers presented in this collection range from descriptions of software tools supporting ontology development and annotation of objects with ontology terms, to applications of text mining for structured relation extraction involving diseases and phenotypes, to detailed proposals for new ontologies and mapping of existing ontologies. Together, the papers consider a range of representational issues in bio-ontology development, and demonstrate the applicability of bio-ontologies to support biological and clinical knowledge-based decision making and analysis.

The full set of papers in the Thematic Issue is available at http://www.biomedcentral.com/collections/sig.

SBOL Visual: A Graphical Language for Genetic Designs

Synthetic Biology Open Language (SBOL) Visual is a graphical standard for genetic engineering. It consists of symbols representing DNA subsequences, including regulatory elements and DNA assembly features. These symbols can be used to draw illustrations for communication and instruction, and as image assets for computer-aided design. SBOL Visual is a community standard, freely available for personal, academic, and commercial use (Creative Commons CC0 license). We provide prototypical symbol images that have been used in scientific publications and software tools. We encourage users to use and modify them freely, and to join the SBOL Visual community: http://www.sbolstandard.org/visual.

Crystal structure of hyaluronidase, a major allergen of bee venom

BACKGROUND

Hyaluronic acid (HA) is the most abundant glycosaminoglycan of vertebrate extracellular spaces and is specifically degraded by a beta-1,4 glycosidase. Bee venom hyaluronidase (Hya) shares 30% sequence identity with human hyaluronidases, which are involved in fertilization and the turnover of HA. On the basis of sequence similarity, mammalian enzymes and Hya are assigned to glycosidase family 56 for which no structure has been reported yet.

RESULTS

The crystal structure of recombinant (Baculovirus) Hya was determined at 1.6 A resolution. The overall topology resembles a classical (beta/alpha)(8) TIM barrel except that the barrel is composed of only seven strands. A long substrate binding groove extends across the C-terminal end of the barrel. Cocrystallization with a substrate analog revealed the presence of a HA tetramer bound to subsites -4 to -1 and distortion of the -1 sugar.

CONCLUSIONS

The structure of the complex strongly suggest an acid-base catalytic mechanism, in which Glu113 acts as the proton donor and the N-acetyl group of the substrate is the nucleophile. The location of the catalytic residues shows striking similarity to bacterial chitinase which also operates via a substrate-assisted mechanism.

Yellow jacket venom allergens, hyaluronidase and phospholipase: sequence similarity and antigenic cross-reactivity with their hornet and wasp homologs and possible implications for clinical allergy

Three known allergens of yellow jacket (Vespula vulgaris) venom are antigen 5, hyaluronidase, and phospholipase. Yellow jacket antigen 5 has been previously cloned and expressed in bacteria; it contains 204 amino acid residues, and it has 69% and 60% sequence identities with the homologous proteins of white-faced hornet (Dolichovespula maculata) and wasp (Polistes annularis), respectively. These studies are now extended to yellow jacket hyaluronidase and phospholipase; they contain 331 and 300 amino acid residues, respectively, and they show 92% and 67% sequence identity with their homologs of white-faced hornet. Tests with the natural and the recombinant vespid allergens in mice indicate partial antigenic cross-reactivity of their homologous proteins at both B- and T-cell levels. There is greater cross-reactivity among hornet and yellow jacket allergens than that among hornet or yellow jacket and wasp allergens. The order of cross-reaction of the three vespid allergens is hyaluronidase > antigen 5 > phospholipase. The continuous (linear) B-cell epitopes of vespid allergens show greater cross-reactivity than their discontinuous epitopes do. The discontinuous B-cell epitopes are immunodominant for all vespid allergens. The low degree of cross-reactivity of the immunodominant discontinuous B-cell epitopes of vespid allergens should be taken into consideration in selection of venoms for immunotherapy of patients with sensitivity to multiple vespids.

Sequence similarity of a hornet (D. maculata) venom allergen phospholipase A1 with mammalian lipases

We have determined the sequence of a venom allergen phospholipase A1 from white-faced hornet (Dolichovespula maculata) by cDNA and protein sequencings. This protein of 300 amino acid residues (Dol m I) has no sequence similarity with other known phospholipases. But it has sequence similarity with mammalian lipases; about 40% identity in overlaps of 123 residues. Tests suggest that hornet phospholipase has weak lipase activity. Hornet venom has 3 major allergens, and another hornet allergen antigen 5 (Dol m V) was previously found to have sequence similarity with a mammalian testis protein and a plant leaf protein.