GWAS Central: an expanding resource for finding and visualising genotype and phenotype data from genome-wide association studies

The GWAS Central resource gathers and curates extensive summary-level genome-wide association study (GWAS) data and puts a range of user-friendly but powerful website tools for the comparison and visualisation of GWAS data at the fingertips of researchers. Through our continued efforts to harmonise and import data received from GWAS authors and consortia, and data sets actively collected from public sources, the database now contains over 72.5 million P-values for over 5000 studies testing over 7.4 million unique genetic markers investigating over 1700 unique phenotypes. Here, we describe an update to integrate this extensive data collection with mouse disease model data to support insights into the functional impact of human genetic variation. GWAS Central has expanded to include mouse gene-phenotype associations observed during mouse gene knockout screens. To allow similar cross-species phenotypes to be compared, terms from mammalian and human phenotype ontologies have been mapped. New interactive interfaces to find, correlate and view human and mouse genotype-phenotype associations are included in the website toolkit. Additionally, the integrated browser for interrogating multiple association data sets has been updated and a GA4GH Beacon API endpoint has been added for discovering variants tested in GWAS. The GWAS Central resource is accessible at https://www.gwascentral.org/.

Auto-CORPus: A Natural Language Processing Tool for Standardizing and Reusing Biomedical Literature

To analyse large corpora using machine learning and other Natural Language Processing (NLP) algorithms, the corpora need to be standardized. The BioC format is a community-driven simple data structure for sharing text and annotations, however there is limited access to biomedical literature in BioC format and a lack of bioinformatics tools to convert online publication HTML formats to BioC. We present Auto-CORPus (Automated pipeline for Consistent Outputs from Research Publications), a novel NLP tool for the standardization and conversion of publication HTML and table image files to three convenient machine-interpretable outputs to support biomedical text analytics. Firstly, Auto-CORPus can be configured to convert HTML from various publication sources to BioC. To standardize the description of heterogenous publication sections, the Information Artifact Ontology is used to annotate each section within the BioC output. Secondly, Auto-CORPus transforms publication tables to a JSON format to store, exchange and annotate table data between text analytics systems. The BioC specification does not include a data structure for representing publication table data, so we present a JSON format for sharing table content and metadata. Inline tables within full-text HTML files and linked tables within separate HTML files are processed and converted to machine-interpretable table JSON format. Finally, Auto-CORPus extracts abbreviations declared within publication text and provides an abbreviations JSON output that relates an abbreviation with the full definition. This abbreviation collection supports text mining tasks such as named entity recognition by including abbreviations unique to individual publications that are not contained within standard bio-ontologies and dictionaries. The Auto-CORPus package is freely available with detailed instructions from GitHub at: https://github.com/omicsNLP/Auto-CORPus.

GWAS Central: a comprehensive resource for the discovery and comparison of genotype and phenotype data from genome-wide association studies

The GWAS Central resource provides a toolkit for integrative access and visualization of a uniquely extensive collection of genome-wide association study data, while ensuring safe open access to prevent research participant identification. GWAS Central is the world's most comprehensive openly accessible repository of summary-level GWAS association information, providing over 70 million P-values for over 3800 studies investigating over 1400 unique phenotypes. The database content comprises direct submissions received from GWAS authors and consortia, in addition to actively gathered data sets from various public sources. GWAS data are discoverable from the perspective of genetic markers, genes, genome regions or phenotypes, via graphical visualizations and detailed downloadable data reports. Tested genetic markers and relevant genomic features can be visually interrogated across up to sixteen multiple association data sets in a single view using the integrated genome browser. The semantic standardization of phenotype descriptions with Medical Subject Headings and the Human Phenotype Ontology allows the precise identification of genetic variants associated with diseases, phenotypes and traits of interest. Harmonization of the phenotype descriptions used across several GWAS-related resources has extended the phenotype search capabilities to enable cross-database study discovery using a range of ontologies. GWAS Central is updated regularly and available at https://www.gwascentral.org.

Factors associated with smoking in low-income pregnant women: relationship to birth weight, stressful life events, social support, health behaviors and mental distress

Since low-income women are at increased risk of having low birth weight infants, factors associated with birth weight among such groups have special relevance. Cigarette-smoking has emerged as an important predictor of low birth weight due to intrauterine growth retardation and pre-term delivery. After confirming the relation of smoking with birth weight, we examined the association of smoking with sociodemographic factors, attitudes towards pregnancy, health behaviors, stressful life events, social support, and symptoms of mental distress in a cohort of 458 Central Harlem women. We found that social support, stress and mental health were associated with smoking behavior but not directly with birth weight. These findings suggest that programs designed to modify health behaviors such as smoking during pregnancy must also take into account such characteristics of the women and their environments which may make behavioral change difficult. Moreover, programs aimed at fostering better health behaviors to improve pregnancy outcome may have to extend beyond the current pregnancy, as indicated by an association between prior adverse pregnancy outcome and smoking in the current pregnancy.

Factors associated with maternal rating of infant health in central Harlem

Maternal perceptions of infant health are strongly associated with the use of health services and may influence the child's future health and development. The extent to which such perceptions reflect infant health problems or a response to environmental stress in the mother's life is unclear. Because this issue may be especially important in disadvantaged groups for which both high levels of infant morbidity and environmental stress are common, we have explored the correlates of maternal perceptions of infant health among 367 mothers of 12-month-olds in central Harlem. Lower ratings of infant health were associated with such indicators of morbidity as problems in the neonatal period, hospitalization, and one or more episodes of illness. They were also associated with lower family income and the mother's rating of her own health in the bivariate analyses, but not in multivariate analyses. Neither maternal mental health, social support, nor environmental stress appeared to influence maternal rating of child health. These findings indicate that a mother's ratings of her infant's health relate specifically to the child's morbidity, independent of her own health and environmental stress. Moreover, within disadvantaged communities, infants of the poorest mothers may experience greater levels of morbidity.

Outreach as case finding: its effect on enrollment in prenatal care

One mechanism proposed for encouraging use of prenatal services in low-income communities involves the employment of community residents to encourage and support individuals in the use of available medical care. This study examines the effect of such workers on the start of prenatal care among a cohort of women receiving prenatal services at clinics affiliated with Harlem Hospital. Of the 599 women enrolling for prenatal care during the intake period for the study, only 52 had had an outreach contact before the start of prenatal care despite extensive field activity. No difference in the week of gestation at the start of prenatal care was noted between those with and without outreach contacts. Controlling for factors associated both with trimester of initiation of prenatal care in this population and with outreach contact did not account for this lack of difference. This type of outreach, known as case-finding, proved to be very labor intensive. Other less costly techniques for encouraging use of services in mobile, urban populations should be sought.

Outreach as case finding. The process of locating low-income pregnant women

This article documents the process of an outreach program for locating disadvantaged women who, as a group, do not receive prenatal care early and have adverse pregnancy outcomes. Three full-time community residents searched for pregnant women for a year, being paid a commission for each woman that they found who enrolled for antenatal care. Outreach workers spent more than half of their time in the field, contacted 20 to 25 people per day, and used a variety of strategies to locate women. Fifty-two women entered the Harlem Hospital Medical Center health care system through the outreach process, with 104 pregnant women not already receiving antenatal care being identified by the outreach workers. Their effort, conversion rate, and yield were comparable to private sector salespeople. The cost per enrollee was high (although not higher than the cost of additional low-birthweight births). Alternatives for locating pregnant women are suggested.

The planning of pregnancy among low-income women in central Harlem

A planned pregnancy is considered desirable, in part because of the potential of a better pregnancy outcome. Since the improvement of pregnancy outcome is of particular relevance in low-income populations, we have compared the characteristics of women with planned and unplanned pregnancies in central Harlem with regard to those factors that might affect pregnancy planning such as sociodemographic factors, attitudes toward child-rearing, environmental stress, social support, and maternal mental health. Of the 416 women in the study, a minority (27%) reported their pregnancy as being planned. They differed from the remainder in being more likely to be married and/or living with a boyfriend or husband and to have been born outside New York City. The two groups did not differ in any other risk factor or in outcome in terms of birth weight and gestational age. The results provide little support for the lack of planning of pregnancy as an indicator of risk in a low-income population and suggest that improvement of perinatal outcome must involve more broadly based interventions that are not confined to the periconceptional period.