Cheiron is built on the literature pharma and biotech teams already trust: including peer-reviewed publications, regulatory filings, clinical trial registries, patent literature, international harmonized guidelines, and your own organization's documents.

Cheiron searches across PubMed, the world's largest biomedical citation database, with over 37 million records spanning life sciences, clinical medicine, and translational research. Cheiron also incorporates curated peer-reviewed journals beyond PubMed's index where they're relevant to a question.

For open-access articles, Cheiron reasons over the full text. For paywalled articles, it uses what's publicly available, typically the abstract. To go beyond the abstract on a paywalled paper, upload the full article as an attachment, into a Project, or into Company Knowledge — Cheiron will then analyze it as a first-class source.

Cheiron pulls from ClinicalTrials.gov, which hosts over 500,000 registered studies worldwide. Cheiron reads structured trial fields — sponsor, phase, status, primary and secondary endpoints, enrollment, posted results — so it can answer questions about trial design, evidence, and competitive landscape with precision.

Cheiron indexes regulatory documents directly from the major authorities its users work with every day:

For most regulatory documents, Cheiron reasons over the full text.

Cheiron searches USPTO for US patent applications and granted patents. This opens IP-driven workflows — patent landscapes, prior-art searches, freedom-to-operate analyses, and competitive pipeline scouting. Combined with regulatory and clinical-trial data, it gives BD and IP teams a fuller picture of competitive activity than any single source can provide on its own.

Internal documents you upload — research papers, CSRs, internal memos, regulatory drafts, competitive intel — become first-class sources alongside the public corpus. Cheiron searches and cites across them with the same rigor, and Deep Citations works on them the same way it works on a PubMed paper. Documents stay in your account and are not used to train models. See the Projects guide and Company Knowledge guide for how to bring your team's materials in.

For enterprise customers, Cheiron can connect directly to proprietary internal data sources — internal databases, document repositories, knowledge bases — so your team works against an integrated corpus without per-document uploads. Contact sales to discuss what your team needs.

Cheiron defaults to its trusted indexed sources for every question. When the information you need sits outside those databases — non-English regulatory guidance from authorities in China, Japan, Vietnam, or Latin America; professional society guidelines from ASCO, ESMO, ASH and similar bodies; or breaking news on approvals, safety signals, and pipeline activity — Cheiron extends its search to the open web.

Web sources go through the same Deep Citation verification as everything else. You can trace any claim back to its original source regardless of where Cheiron found it. This is what makes Cheiron meaningfully more global than tools restricted to a fixed set of databases, without giving up verifiability.

Cheiron refreshes its corpus on rolling cycles. Most sources are updated daily; newer publications and filings typically appear within one to two days. Web search runs against the live web, so anything currently published online is reachable through Cheiron's pipeline immediately.

Cheiron's corpus expands often, so recent additions may land in the product before this article catches up. For the most current state on a specific source, or to request a new integration, write to support@cheiron.bio.