Biology Atlas

Map / Concepts

concept · intermediate · partly written

The QIAGEN Knowledge Base (behind Ingenuity Pathway Analysis)

QIAGEN's proprietary, manually curated knowledge base of literature findings with experimental context, sold inside IPA and as the QIAGEN Biomedical Knowledge Base.

Also called: QIAGEN Knowledge Base, Ingenuity Knowledge Base, QIAGEN Biomedical Knowledge Base, KB-HD, KB-AI, Kaijen (spelling in the original request)

Short answer

Two products share one curated core: the QIAGEN Knowledge Base inside Ingenuity Pathway Analysis, and the QIAGEN Biomedical Knowledge Base sold as data in two tiers, KB-HD (human-curated) and KB-AI (AI-extracted). Every size and curation figure is QIAGEN's own claim, and no outside accuracy audit was found. 123

Reported, not yet replicated

History more

QIAGEN's filing reports buying Ingenuity Systems for $107.0 million in cash with $10.0 million in escrow, dated 29 April 2013 in one section and 30 April in another. KB-HD launched on 25 April 2022 and KB-AI on 27 February 2024, both by company press release. 452

Reported, not yet replicated

How it is built (company's claim) more

The company says the data are manually curated by MD- and PhD-certified experts in a proprietary ontology, with synonyms mapped to public identifiers; an interview with a company product manager cites over 150 experts and 25 years. Which journals are read, how curators' conflicts are settled and error rates are not stated anywhere opened. 16

Reported, not yet replicated

Sizes, with dates more

In 2013 the causal network built from nearly 5 million findings had about 40,000 nodes and 1,480,000 edges (a paper by company staff). The 2026 trainer slide says 14+ million findings; the KB-HD page says 24 million relationships and over 13 million findings; the 2024 release says KB-AI holds 640 million relationships, over 600 million more than KB-HD. Node-type and edge-type counts were not found. 7812

Reported, not yet replicated

What a finding carries, and what context is recorded more

A trainer's slide describes a finding with type, direction, effect (increases or decreases), directness, and any of tissue, location, cell line, organism, experiment and a source PMID. Context is recorded when curated, not uniformly: one example lists tissue and location but no organism, another cell line, organism and experiment. Several findings can exist for one pair. No evidence sentence was shown on any page opened and a per-finding confidence was not found; the 2013 paper gives edge weights reflecting confidence in direction. 87

Reported, not yet replicated

Context was pooled in the published causal algorithm more

The 2013 paper, by company staff, did not filter by context: mouse, rat and human findings were pooled because filtering made networks too sparse, and context filtering was listed as future work. Whether current IPA filters by context was not checked. 7

Reported, not yet replicated

What outsiders have measured more

No head-to-head with Reactome, STRING, Hetionet or INDRA-style assembly, no curation audit and no test of KB-AI was found. A 2010 outside study found low consistency between KEGG, Ingenuity and WikiPathways (Ingenuity-KEGG apoptosis gene overlap 33 to 37%, pair overlap 12 to 14%), which measures disagreement on a 2010 version, not accuracy. A 2011 benchmark against transcription-factor targets found overlap for most pathway databases surprisingly small (abstract only; the IPA-specific result is unverified). An IPA analysis could not be exactly replicated because the gene universe is updated often. The company's own 2013 validation ranked TNF first in 8 of 10 datasets. 91011127

Reported, not yet replicated

Licensing and access as publicly stated deep

KB-HD is described as direct access with API access to the full data or subsets and updates during a subscription; file formats, price and licence terms are not stated on the page. IPA is licensed software. The licence agreements could not be read, so what may be exported or republished is unverified. 13

Unverified: from our own report, source not opened

How it differs in philosophy deep

PrimeKG, Hetionet and SPOKE collapse many databases into one typed relation. INDRA computes belief from the number and kind of evidence and publishes accuracy on a curated sample (about 73%). QIAGEN is nearest INDRA's assertion-plus-evidence end but works by human curation and asserts trust by employing curators; it keeps the richest record of context and the weakest public proof of quality. This reading is our own comparison of the sources. 1713

Our inference

Still open here

Not yet written: Plain answer, Why it matters, Equation, Real numbers.

Linked from

Sources

  1. QIAGEN. QIAGEN Biomedical Knowledge Base product page. qiagen.com (2026). vendor-page, not peer reviewed; opened 2026-11-18 all records citing this
  2. QIAGEN. QIAGEN launches AI-derived biomedical knowledge base to accelerate data-driven drug discovery. corporate.qiagen.com (2024). organisation-page, not peer reviewed; opened 2026-11-18 all records citing this
  3. QIAGEN. Ingenuity Pathway Analysis product page. qiagen.com (2026). vendor-page, not peer reviewed; opened 2026-11-18 all records citing this
  4. QIAGEN N.V.. QIAGEN Form 6-K reporting the acquisition of Ingenuity Systems. SEC EDGAR (2013). regulatory, not peer reviewed; opened 2026-11-18 all records citing this
  5. QIAGEN. QIAGEN gives drug research direct access to world-leading data with launch of QIAGEN Biomedical Knowledge Base. corporate.qiagen.com (2022). organisation-page, not peer reviewed; opened 2026-11-18 all records citing this
  6. SiliconANGLE. QIAGEN grounds AI drug discovery agents in curated knowledge. siliconangle.com (2026). news, not peer reviewed; opened 2026-11-18 all records citing this
  7. Krämer A, et al.. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics (2013). paper; opened 2026-11-18 all records citing this
  8. QIAGEN trainer, NCI BTEP. IPA training slides, NCI Bioinformatics Training and Education Program. bioinformatics.ccr.cancer.gov (2026). web, not peer reviewed; opened 2026-11-18 all records citing this
  9. Soh D, Dong D, Guo Y, Wong L. Consistency, comprehensiveness, and compatibility of pathway databases. BMC Bioinformatics (2010). paper; opened 2026-11-18 all records citing this
  10. Shmelkov E, et al.. Assessing quality and completeness of human transcriptional regulatory pathways on a genome-wide scale. Biology Direct (2011). paper; opened 2026-11-18 all records citing this
  11. Henderson-MacLennan NK, et al.. Pathway analysis software: annotation errors and solutions. Molecular Genetics and Metabolism (2010). paper; opened 2026-11-18 all records citing this
  12. Timmons JA, Szkop KJ, Gallagher IJ. Multiple sources of bias confound functional enrichment analysis of global -omics data. Genome Biology (2015). paper; opened 2026-11-18 all records citing this
  13. INDRA paper on automated assembly of mechanistic knowledge from text and databases (PMC10167483). PubMed Central (2023). paper; opened 2026-11-17 all records citing this
Where it sits on the map
History of this record (6)

This record has been changed 6 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.

Status
partly written
Last updated
2026-11-19
Record
content/concepts/qiagen-knowledge-base.json