Gene annotation bias impedes biomedical research.
Ontology highlight
ABSTRACT:
SUBMITTER: Haynes WA
PROVIDER: S-EPMC5778030 | biostudies-literature | 2018 Jan
REPOSITORIES: biostudies-literature
Publications
Gene annotation bias impedes biomedical research.
Haynes Winston A WA Tomczak Aurelie A Khatri Purvesh P
Scientific reports 20180122 1
We found tremendous inequality across gene and protein annotation resources. We observed that this bias leads biomedical researchers to focus on richly annotated genes instead of those with the strongest molecular data. We advocate that researchers reduce these biases by pursuing data-driven hypotheses. ...[more]
Similar Datasets
Project description:Female mammals have long been neglected in biomedical research. The NIH mandated enrollment of women in human clinical trials in 1993, but no similar initiatives exist to foster research on female animals. We reviewed sex bias in research on mammals in 10 biological fields for 2009 and their historical precedents. Male bias was evident in 8 disciplines and most prominent in neuroscience, with single-sex studies of male animals outnumbering those of females 5.5 to 1. In the past half-century, male bias in non-human studies has increased while declining in human studies. Studies of both sexes frequently fail to analyze results by sex. Underrepresentation of females in animal models of disease is also commonplace, and our understanding of female biology is compromised by these deficiencies. The majority of articles in several journals are conducted on rats and mice to the exclusion of other useful animal models. The belief that non-human female mammals are intrinsically more variable than males and too troublesome for routine inclusion in research protocols is without foundation. We recommend that when only one sex is studied, this should be indicated in article titles, and that funding agencies favor proposals that investigate both sexes and analyze data by sex.
| S-EPMC3008499 | biostudies-literature
Project description:In animal-based biomedical research, both the sex and the age of the animals studied affect disease phenotypes by modifying their susceptibility, presentation and response to treatment. The accurate reporting of experimental methods and materials, including the sex and age of animals, is essential so that other researchers can build on the results of such studies. Here we use text mining to study 15,311 research papers in which mice were the focus of the study. We find that the percentage of papers reporting the sex and age of mice has increased over the past two decades: however, only about 50% of the papers published in 2014 reported these two variables. We also compared the quality of reporting in six preclinical research areas and found evidence for different levels of sex-bias in these areas: the strongest male-bias was observed in cardiovascular disease models and the strongest female-bias was found in infectious disease models. These results demonstrate the ability of text mining to contribute to the ongoing debate about the reproducibility of research, and confirm the need to continue efforts to improve the reporting of experimental methods and materials.
| S-EPMC4821800 | biostudies-literature
Project description:Gene function annotation remains a key challenge in modern biology. This is especially true for high-throughput techniques such as gene expression experiments. Vital information about genes is available electronically from biomedical literature in the form of full texts and abstracts. In addition, various publicly available databases (such as GenBank, Gene Ontology and Entrez) provide access to gene-related information at different levels of biological organization, granularity and data format. This information is being used to assess and interpret the results from high-throughput experiments. To improve keyword extraction for annotational clustering and other types of analyses, we have developed a novel text mining approach, which is based on keywords identified at the level of gene annotation sentences (in particular sentences characterizing biological function) instead of entire abstracts. Further, to improve the expressiveness and usefulness of gene annotation terms, we investigated the combination of sentence-level keywords with terms from the Medical Subject Headings (MeSH) and Gene Ontology (GO) resources. We find that sentence-level keywords combined with MeSH terms outperforms the typical 'baseline' set-up (term frequencies at the level of abstracts) by a significant margin, whereas the addition of GO terms improves matters only marginally. We validated our approach on the basis of a manually annotated corpus of 200 abstracts generated on the basis of 2 cancer categories and 10 genes per category. We applied the method in the context of three sets of differentially expressed genes obtained from pediatric brain tumor samples. This analysis suggests novel interpretations of discovered gene expression patterns.
| S-EPMC2241933 | biostudies-literature
Project description:Progress toward gender equality is thwarted by the underrepresentation of women in political leadership, even as most Americans report they would vote for women candidates. Here, we hypothesize that women candidates are often disadvantaged by pragmatic bias, a tendency to withhold support for members of groups for whom success is perceived to be difficult or impossible to achieve. Across six studies (N = 7,895), we test whether pragmatic bias impedes women's access to a highly significant political leadership position-the US presidency. In two surveys, 2020 Democratic primary voters perceived women candidates to be less electable, and these beliefs were correlated with lower intentions to vote for women candidates (Studies 1 and 2). Voters identified many reasons women would be less electable than men, including others' unwillingness to vote for women, biased media coverage, and higher requirements to prove themselves. We next tested interventions to reduce pragmatic bias. Merely correcting misperceptions of Americans' reported readiness for a woman president did not increase intentions to vote for a woman (Study 3). However, across three experiments (including one preregistered on a nationally representative sample), presenting evidence that women earn as much support as men in US general elections increased Democratic primary voters' intentions to vote for women presidential candidates, an effect driven by heightened perceptions of these candidates' electability (Studies 4 to 6). These findings highlight that social change efforts can be thwarted by people's sense of what is possible, but this may be overcome by credibly signaling others' willingness to act collectively.
| S-EPMC8833189 | biostudies-literature
Project description:Systematic under-representation of pregnant women and gender diverse pregnant people in clinical research has prevented them from benefitting fairly from biomedical advances. The resulting lack of pharmacological safety and efficacy data leads to medicine discontinuation, sub-optimal dosing, and reliance on repurposed therapies. We identify four roadblocks to fair inclusion. First, investment and research are inhibited by protectionist attitudes among research gatekeepers who view pregnancy as a vulnerable state. Second, exclusion ignores human-specific biological variations affecting medication absorption and impacts on the pregnant body. Third, pregnant populations in low-and middle-income countries face a double disadvantage due to gender and location, despite bearing a disproportionate maternal mortality burden. Fourth, perspectives and experiences of pregnant populations are undervalued in clinical intervention design. We propose five actions to optimize fair inclusion: fostering reciprocal partnerships, prioritizing multi-disciplinary research, awareness-raising of the need for pharmaceutical innovation, conducting regulatory analyses, and promoting responsible inclusion over presumptive exclusion.
| S-EPMC11500090 | biostudies-literature
Project description:A major challenge in microarray data analysis is the functional interpretation of gene lists. A common approach to address this is over-representation analysis (ORA), which uses the hypergeometric test (or its variants) to evaluate whether a particular functionally defined group of genes is represented more than expected by chance within a gene list. Existing applications of ORA have been largely limited to pre-defined terminologies such as GO and KEGG. We report our explorations of whether ORA can be applied to a wider mining of free-text. We found that a hitherto underappreciated feature of experimentally derived gene lists is that the constituents have substantially more annotation associated with them, as they have been researched upon for a longer period of time. This bias, a result of patterns of research activity within the biomedical community, is a major problem for classical hypergeometric test-based ORA approaches, which cannot account for such bias. We have therefore developed three approaches to overcome this bias, and demonstrate their usability in a wide range of published datasets covering different species. A comparison with existing tools that use GO terms suggests that mining PubMed abstracts can reveal additional biological insight that may not be possible by mining pre-defined ontologies alone.
| S-EPMC2699530 | biostudies-literature
Project description:The rabbit is a classic animal model for biomedical research, but the production of gene targeted transgenic rabbits had been extremely challenging until the recent advent of gene editing tools. More than fifty gene knockout or knock-in rabbit models have been reported in the past decade. Gene edited (GE) rabbit models, compared to their counterpart mouse models, may offer unique opportunities in translational biomedical research attributed primarily to their relatively large size and long lifespan. More importantly, GE rabbit models have been found to mimic several disease pathologies better than their mouse counterparts particularly in fields focused on genetically inherited diseases, cardiovascular diseases, ocular diseases, and others. In this review we present selected examples of research areas where GE rabbit models are expected to make immediate contributions to the understanding of the pathophysiology of human disease, and support the development of novel therapeutics.
| S-EPMC7876448 | biostudies-literature
Project description:Recent technological advances have resulted in an unprecedented increase in publicly available biomedical data, yet the reuse of the data is often precluded by experimental bias and a lack of annotation depth and consistency. Missing annotations makes it impossible for researchers to find datasets specific to their needs. Here, we investigate RNA-sequencing metadata prediction based on gene expression values. We present a deep-learning-based domain adaptation algorithm for the automatic annotation of RNA-sequencing metadata. We show, in multiple experiments, that our model is better at integrating heterogeneous training data compared with existing linear regression-based approaches, resulting in improved tissue type classification. By using a model architecture similar to Siamese networks, the algorithm can learn biases from datasets with few samples. Using our novel domain adaptation approach, we achieved metadata annotation accuracies up to 15.7% better than a previously published method. Using the best model, we provide a list of >10,000 novel tissue and sex label annotations for 8,495 unique SRA samples. Our approach has the potential to revive idle datasets by automated annotation making them more searchable.
| S-EPMC8559615 | biostudies-literature
Project description:BackgroundLarge-scale genomic studies based on transcriptome technologies provide clusters of genes that need to be functionally annotated. The Gene Ontology (GO) implements a controlled vocabulary organised into three hierarchies: cellular components, molecular functions and biological processes. This terminology allows a coherent and consistent description of the knowledge about gene functions. The GO terms related to genes come primarily from semi-automatic annotations made by trained biologists (annotation based on evidence) or text-mining of the published scientific literature (literature profiling).ResultsWe report an original functional annotation method based on a combination of evidence and literature that overcomes the weaknesses and the limitations of each approach. It relies on the Gene Ontology Annotation database (GOA Human) and the PubGene biomedical literature index. We support these annotations with statistically associated GO terms and retrieve associative relations across the three GO hierarchies to emphasise the major pathways involved by a gene cluster. Both annotation methods and associative relations were quantitatively evaluated with a reference set of 7397 genes and a multi-cluster study of 14 clusters. We also validated the biological appropriateness of our hybrid method with the annotation of a single gene (cdc2) and that of a down-regulated cluster of 37 genes identified by a transcriptome study of an in vitro enterocyte differentiation model (CaCo-2 cells).ConclusionThe combination of both approaches is more informative than either separate approach: literature mining can enrich an annotation based only on evidence. Text-mining of the literature can also find valuable associated MEDLINE references that confirm the relevance of the annotation. Eventually, GO terms networks can be built with associative relations in order to highlight cooperative and competitive pathways and their connected molecular functions.
| S-EPMC1482722 | biostudies-literature
Project description:Computational annotation of textual information has taken on an important role in knowledge extraction from the biomedical literature, since most of the relevant information from scientific findings is still maintained in text format. In this endeavour, annotation tools can assist in the identification of biomedical concepts and their relationships, providing faster reading and curation processes, with reduced costs. However, the separate usage of distinct annotation systems results in highly heterogeneous data, as it is difficult to efficiently combine and exchange this valuable asset. Moreover, despite the existence of several annotation formats, there is no unified way to integrate miscellaneous annotation outcomes into a reusable, sharable and searchable structure. Taking up this challenge, we present a modular architecture for textual information integration using semantic web features and services. The solution described allows the migration of curation data into a common model, providing a suitable transition process in which multiple annotation data can be integrated and enriched, with the possibility of being shared, compared and reused across semantic knowledge bases.
| S-EPMC5691355 | biostudies-literature