THEOBROMA: an aggregated open database of 1.13 million natural products with per-compound license auditing, three-tier classification, and stereochemistry-aware deduplication
THEOBROMA is an open database aggregating over 1.13 million natural products from 29 sources that distinguishes itself through per-compound license auditing, a three-tier classification system, and stereochemistry-aware deduplication to enable license-compliant virtual screening and isomer-specific bioactivity analysis.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Imagine the world of drug discovery as a massive, bustling library where scientists are hunting for rare, magical ingredients found in nature (like plants and fungi) to create new medicines. For a long time, this library has been a bit chaotic. Different sections of the library had different rules about who could borrow books, and the librarians often treated the entire section as having one rule, even if individual books had different permissions.
THEOBROMA is a new, super-organized digital catalog that fixes these problems. Here is how it works, broken down into simple parts:
1. The "License Detective" (Per-Compound Auditing)
In the old system, if one book in a section had a strict "Do Not Copy" rule, the whole section was treated as off-limits. This caused a problem recently when a major source changed its rules, suddenly locking up thousands of useful ingredients that scientists thought were free to use.
THEOBROMA acts like a super-detective that checks the legal paperwork for every single ingredient individually, not just the whole shelf.
- The Rule: It uses a "strictest rule wins" approach. If a chemical is listed in three different places, and two say "Free to use" but one says "Restricted," THEOBROMA marks that specific chemical as "Restricted."
- The Result: It found over 900,000 ingredients (about 79%) that are truly free for anyone to use. It also keeps a clear receipt (an "audit trail") showing exactly where the permission rules came from, so scientists know they are safe to use them.
2. The "Confidence Ladder" (Three-Tier Classification)
Not all information in the library is equally reliable. Some books are written by experts who double-checked the facts, while others are just rough notes or guesses.
THEOBROMA organizes its 1.13 million ingredients into a three-step ladder so scientists know how much to trust the data:
- Top Tier (Curated): 35% of the items are like gold-standard textbooks, carefully checked by human experts.
- Middle Tier (High-Confidence Inferred): 44% are like well-researched articles where the facts are strongly supported by evidence but weren't manually checked by a human.
- Bottom Tier (Exploratory): 10% are like "wild guesses" or early drafts. They are still useful for exploring new ideas, but scientists know to be extra careful.
3. The "Twin Tracker" (Stereochemistry Deduplication)
In chemistry, some ingredients are like left and right hands. They look identical on paper, but they are mirror images of each other. In the real world, one might be a cure, while its mirror image might be useless or even harmful.
Old databases often treated these "mirror twins" as the exact same thing, mixing them up. THEOBROMA is like a sharp-eyed twin tracker. It uses a special 27-character code (a unique fingerprint) to tell these mirror images apart.
- It groups these twins into "families" so scientists can see them clearly.
- It provides a special tool to look at these families, ensuring that when a scientist studies a specific shape of a molecule, they aren't accidentally studying its mirror image by mistake.
The Bottom Line
THEOBROMA is a massive, open database containing over 1.13 million natural ingredients. Its main superpower is that it doesn't just dump data; it audits the legal rights of every single item, grades the reliability of the information, and separates mirror-image twins that other databases mix together. This allows scientists to search through the entire collection safely, knowing exactly what they can use, how much they can trust it, and which specific shape of the molecule they are looking at.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.