HumanMine: advanced data searching, analysis and cross-species comparison

Rachel Lyne, Adrián Bazaga, Daniela Butano, Sergio Contrino, Joshua Heimbach, Fengyuan Hu, Alexis Kalderimis, Mike Lyne, Kevin Reierskog, Radek Štěpán, Julie Sullivan, Archie Wise, Yo Yehudi, Gos Micklem

TL;DR

HumanMine is a powerful, integrated database for human genomics and proteomics data, built on the InterMine platform. It offers a user-friendly web interface and a scriptable API for sophisticated data exploration, analysis, and cross-species comparison, enabling researchers to uncover new biological relationships.

What Problem Does This Paper Address?

The explosion of biological data from genomics, proteomics, and other high-throughput experiments presents a significant challenge for researchers. Integrating and analyzing these diverse datasets from multiple sources is complex and time-consuming, often requiring specialized bioinformatics skills and the use of numerous disparate databases. This fragmentation hinders the ability to perform integrative, multi-staged analyses that are crucial for understanding complex biological systems and uncovering novel insights. HumanMine aims to solve this by providing a unified platform that brings together a vast array of human biological data, making it accessible and analyzable through intuitive tools.

Methods

HumanMine is built upon the InterMine data integration platform, a robust framework designed for integrating diverse biological datasets. Key features of HumanMine's approach include:

  • Data Integration: Combining data from experimental, computational, and curated sources into a single, searchable database.
  • User-Friendly Interface: A web interface with sophisticated query options, identifier resolution, interactive results tables, and graphical analysis tools for data exploration.
  • Web Services API: A scriptable API for programmatic access to the data, supporting common scripting languages.
  • Integrative Analysis: Facilitating multi-staged analyses by allowing users to work with single items or lists of items, and to perform enrichment analyses on these lists.
  • Cross-Species Comparison: Interoperability with other InterMine databases for model organisms (e.g., mouse, rat, zebrafish, fly, worm, yeast) to enable comparative studies.
  • Template Searches and Query Builder: Pre-defined "canned" searches and a flexible query builder for constructing complex queries.

Datasets

HumanMine v12, as described in the paper, integrates data from approximately 40 different data sets. These include:

  • Human Reference Genome and Annotation: Latest release from NCBI.
  • High-throughput Datasets: Covering genome, transcriptome, interactome, and proteome.
  • Specific Data Types: Full genome and proteome annotation, gene-pathway and Gene Ontology annotations, expression data (including tissue- and disease-specific), proteome-wide and genetic interaction data, disease and phenotype data (including Genome-Wide Association Studies), and clinically significant SNPs and cis-eQTLs.
  • External Data Sources: Data from sources like ArrayExpress, GTex, Human Protein Atlas, OMIM, ClinVar, GWAS Catalogue, DisGenNet, HPO, DrugBank, IntAct, BioGrid, and Reactome are utilized.

Key Insights and Ideas

  • The Power of Integration: HumanMine demonstrates that integrating diverse biological data sources into a single platform significantly enhances the ability to discover complex relationships that would be missed when analyzing data in isolation.
  • Facilitating Complex Biological Questions: The platform is designed to support sophisticated, multi-stage analyses, moving beyond simple data retrieval to enabling hypothesis generation and testing.
  • Bridging the Gap Between Data and Insight: By providing both intuitive graphical interfaces and a powerful API, HumanMine caters to a wide range of users, from bench biologists to computational researchers.
  • Model Organism Data is Key: The interoperability with other InterMine databases highlights the importance of comparative genomics and the use of model organisms to understand human biology.
  • Disease Comorbidity as a Use Case: The paper effectively showcases how HumanMine can be used to explore complex biological questions, such as identifying shared pathways and genes in disease comorbidities like asthma and type 2 diabetes.

Key Results

The paper presents several use cases to illustrate HumanMine's capabilities:

  • PAX6 Gene Exploration: A simple keyword search for PAX6 quickly returns links to genes, publications, interaction data, proteins, and more. The report page for PAX6 provides a comprehensive summary of its function, expression, and associated diseases.
  • List Analysis and Enrichment: Selecting a gene set (e.g., from a Genomics England gene panel) allows for enrichment analysis, revealing statistically over-represented Gene Ontology terms and pathways. For the PAX6 example, this identified enrichment in visual and sensory organ development.
  • Disease Comorbidity Analysis (Asthma and Type 2 Diabetes):
    • A shared set of 912 genes was identified between asthma and type 2 diabetes.
  • Enrichment analysis of this shared set highlighted the immune system, signal transduction, and hemostasis as key biological functions.
  • Specific genes like TGFB1 and Toll-like receptors (TLRs) were identified as potentially important in the comorbidity.
  • Expression data confirmed that many shared genes are expressed in relevant tissues like lung and adipose tissue.
  • Literature searches using HumanMine identified publications linking TLR2 polymorphisms to type 1 diabetes and allergic asthma.

Limitations and Future Work

  • Data Updates: Data are updated quarterly, which is standard but means there can be a lag for the very latest findings.
  • User Interface Enhancements: While the paper mentions a new interface (BlueGenes), it also notes that some features (like overlap notifications between user lists and public lists) are still under development for the user interface.
  • Future Work: The authors suggest that the methods and workflows demonstrated can be applied to any combination of disease comorbidity and other research areas, implying ongoing development and expansion of use cases. The platform's design inherently supports the inclusion of new data sources as they become available.

Summary updated

Abstract

Abstract HumanMine ( www.humanmine.org ) is an integrated database of human genomics and proteomics data that provides a powerful interface to support sophisticated exploration and analysis of data compiled from experimental, computational and curated data sources. Built using the InterMine data integration platform, HumanMine includes genes, proteins, pathways, expression levels, SNPs, diseases and more, integrated into a single searchable database. HumanMine promotes integrative analysis, a powerful approach in modern biology that allows many sources of evidence to be analysed together. The data can be accessed through a user-friendly web interface as well as a powerful, scriptable web service API to allow programmatic access to data. The web interface includes a useful identifier resolution system, sophisticated query options and interactive results tables that enable powerful exploration of data, including data summaries, filtering, browsing and export. A set of graphical analysis tools provide a rich environment for data exploration including statistical enrichment of sets of genes or other biological entities. HumanMine can be used for integrative multi-staged analysis that can lead to new insights and uncover previously unknown relationships. Database URL: https://www.humanmine.org
BibTeX
@article{lyne2022humanmine,
  author = {Rachel Lyne and Adrián Bazaga and Daniela Butano and Sergio Contrino and Joshua Heimbach and Fengyuan Hu and Alexis Kalderimis and Mike Lyne and Kevin Reierskog and Radek Štěpán and Julie Sullivan and Archie Wise and Yo Yehudi and Gos Micklem},
  title = {{HumanMine: advanced data searching, analysis and cross-species comparison}},
  journal = {Database J. Biol. Databases Curation},
  year = {2022},
  doi = {10.1093/database/baac054},
}
Figure