Loading ...

PharmaDive is a dataset covering more than 16,000 substances—drugs, natural compounds, research chemicals, venoms, extracts, and more—each rated on 130 structural, chemical, pharmacological, and clinical attributes. Instead of pulling facts from documents, it has a web-search-enabled language model judge every attribute for every compound on a graded scale from “definitely no” to “definitely yes”. Just five years ago, a project like PharmaDive would have been nearly impossible without a generous, multi-year grant. With today’s A.I. capabilities, I was able to build PharmaDive with about $7,000 in Grok (grok.com) API costs.

The manuscript describing the data and methods is here, and all code for using the pipeline for your own work is here. Tyler can be contacted at tymoore@pennmedicine.upenn.edu.

Download the full data set here.

Suggested citation:
Moore, T. M. (2026). PharmaDive: A Language Model Generates an Uncertainty-Graded Property Matrix for More Than 16,000 Compounds. Retrieved from https://pharmadive.org.

An interface for exploring the database is available below.

Search 16,116 compounds by property: describe the profile you are looking for on 11 dimensions, or start from compounds you already know.

Compounds
16,116
Dimensions
11
Scale
−4 to +4

Search

Two ways to query the resource

Set the dimensions that matter and switch off the rest. Zero is the average compound in the resource; the ends of each scale are its extremes.

Acts on the nervous system.

0.00
From −4, Not neuroactive, to +4, Strongly neuroactive. Zero is the average compound.

Modulates immune function.

0.00
From −4, No immune effect, to +4, Strongly immunomodulatory. Zero is the average compound.

A plant-derived compound or extract.

0.00
From −4, Not plant-derived, to +4, Phytochemical. Zero is the average compound.

A conventional synthetic small-molecule agent rather than a biologic, extract, or natural product.

0.00
From −4, Biologic or natural, to +4, Synthetic small molecule. Zero is the average compound.

Contains a basic, nitrogen-bearing ring system.

0.00
From −4, No basic ring, to +4, Basic heterocycle. Zero is the average compound.

Fat-soluble; crosses membranes readily.

0.00
From −4, Hydrophilic, to +4, Lipophilic. Zero is the average compound.

Influences oxidative stress and redox balance.

0.00
From −4, No redox effect, to +4, Redox-active. Zero is the average compound.

Acts on DNA or RNA.

0.00
From −4, Not nucleic-acid targeting, to +4, Nucleic-acid targeting. Zero is the average compound.

Modifies gene regulation without changing the sequence.

0.00
From −4, No epigenetic effect, to +4, Epigenetic. Zero is the average compound.

Relevant to protection against neurodegeneration.

0.00
From −4, No neuroprotection, to +4, Anti-neurodegenerative. Zero is the average compound.

Acts on the heart, vessels, or metabolism.

0.00
From −4, No cardiometabolic effect, to +4, Cardiometabolic. Zero is the average compound.

Note: Data used for PharmaDive is also accessible via API here.

Other tools from MooreMetrics:
AuthorDive.com
GameDive.org (for video games)
GameDive 1.0 (for all non-video games/sports)
MovieDive.org
PodDive.org (for podcasts)
ShowDive.com
AllDive.org (for all above combined)
SeeDive.org (for location-searchable live webcams)
RecipeDive.com
FashionDive.org
FragranceDive.com
CollegeDive.org
ChowDive.com (for restaurants)
LifeDive.org (for plants, animals, & fungi)
Moore Typology (personality test)

For folks in the Philadelphia area:
PhillyMatch.org
ChowDive Philly

Moore Metrics | Proudly powered by WordPress