Ressources
- Bioschemas Community Initiative: Bioschemas Specifications and Types for Life Sciences Resources, ELIXIR Research Infrastructure. https://bioschemas.org/
- Google Search Central: JavaScript SEO Basics and Server-Side Rendering Guidelines, Google Developers Documentation. https://developers.google.com/search/docs/crawling-indexing/javascript/javascript-seo-basics
- Schema.org Community Group: BioChemEntity and Scientific Dataset Specifications, W3C Web Schemas Group. https://schema.org/BioChemEntity
- National Center for Biotechnology Information (NCBI): Entrez System Architecture and Biomedical Data Indexing, National Library of Medicine (NLM/NIH). https://www.ncbi.nlm.nih.gov/
- World Wide Web Consortium (W3C): Semantic Web and Linked Data in Biomedical Applications, W3C Recommendations. https://www.w3.org/standards/semanticweb/
Life-science websites share a lot of specialized data and proprietary protocols, but search engines seem to have difficulty indexing them. Why?
Standard SEO is essentially built for e-commerce shops or lifestyle blogs and don't work well for scientific contents. When a biology lab or gene bank publishes scientific content on its website, including complex nomenclature, network traffic does not flow smoothly.
We think that technical optimization for search engines, in the biological sciences, requires a customized engineering strategy.
Web teams must manage their contents structures using a dedicated nomenclature adapted for SEO-Engines.
Some technical obstacles (for SEO-engines) found in scientific platforms
Search engine bots visit web pages with a limit of time and resources. Because life sciences websites handle millions of dynamic records and are updated regularly, following research results, for example, it can take time to visit them, this is "the bottlenecks".
A single antibody database might generate billions of filter combinations across host species, target reactivity, clone type, and conjugate options. Search engines can be trapped in these parameter loops. The consequence is that new research pages remain unindexed for an uncertain period of time.
Nomenclature Conflict and Synonym
Biological molecules often have multiple identifiers. A researcher might search for "OCT4", while another searches for "POU5F1", and a third uses "Entrez Gene ID 5460".
Search engines often fail or take time to make a link between these terms. If a website doesn't explicitly declare the relationships between them, there is a risk of missing some web traffic.
Fixing these issues requires some basic changes to your website: the use of canonnical description of the content.
How to manage it?
With the help of implementing “Bioschemas” and “Extended Schema.org”
The standard Schema.org nomenclature covers basic organizations and blog articles, but it is insufficient for scientific resources.
Bioschemas, a community extension built on Schema.org, provides a better nomenclature applicable to the life sciences.
Exemple
BioChemEntity: Defines chemical compounds, genes, and molecular targets.
Protein and Gene: Includes identifiers from UniProt, Ensembl, and NCBI.
Dataset: Details biological repositories, download formats, citation requirements, and licensing.
LabProtocol: Describes experimental workflows, reagents, and equipment specifications.
Furthemore, a sitemap can help search engines to find the relevant data.