By the Numbers
A live snapshot of the library: how much it covers, where it’s deep, and (just as usefully) where it’s thin. The gaps are not hidden: sparse species pages and short coverage bars are the clearest places to contribute. All figures are regenerated from the canonical Markdown on every build.
Research coverage
The matrix maps 27 AI methods against 8 research areas: 78 of 216 cells (36.1%) carry at least one paper. Where the bars are short, the field (and this library) is still thin.
Subject coverage
Every paper, tool, database and dataset is tagged on one shared subject axis of 8 themes and 16 finer tags. That comes to 1,660 tags across 818 of 876 items (93%). A theme counts an item once, however many of its tags apply, so these bars overlap rather than partition. Browse any of them on Topics.
Themes are a different axis from the 8 research areas above, not a renaming of them. Each theme names one research area and each research area is named by one theme, so they pair one-to-one and share a colour. Pairing is not equivalence: the two axes count different populations, so never add or compare the numbers. The matrix classifies what a paper did; themes tag what a resource is about, across papers, software, databases and datasets alike.
Sub-topics
The 16 finer tags under those themes. A tag is minted only once at least three items cluster under it, so this list tracks what the library actually holds. Counts overlap with their theme and with each other.
Most-used methods
Datasets by species: where help is wanted
238 catalogued datasets across 17 pages: 176 per-species inventory rows, 21 curated entries on those same species and cross-species pages (atlases, metabolic models and reference sets), 24 on the reference pages, 17 benchmark datasets. Every species page now has at least one catalogued deposit. The counts below show where coverage is thinnest (Goat, Turkey, Duck carry the fewest), the best place to deepen the library. See Contributing to add a deposit.
Licensing: what you can build on
Every tool, database and curated dataset entry carries a coarse license tier (354 in total; papers are excluded, they carry no license). 142 are classified and 212 are not, so the largest bar below is the work still to do: an unknown tier means nobody has recorded the terms yet, not that the terms are restrictive. Recording one is a small, self-contained contribution; see Contributing, or browse by tier on Licenses. A triage signal only; always confirm at the source.
Licensing by subject
The same 354 entries split by subject and sub-topic, papers excluded. Read the unknown column first: it is the largest in most rows, so the rest of any row is a floor on what is reusable there, not a full picture. Rows overlap, since an item tagged with several subjects appears under each. Any cell opens the resources behind it.
| Subject | Permissive | Copyleft | Restricted | Unknown | All |
|---|---|---|---|---|---|
| AI Methods & Tooling | 38 | 8 | 4 | 101 | 151 |
| AI agents & foundation models | 23 | 2 | 4 | 18 | 47 |
| Benchmarks & evaluation | 1 | 0 | 0 | 25 | 26 |
| Comparative studies | 0 | 0 | 0 | 0 | 0 |
| Bioprocess & Manufacturing | 8 | 3 | 6 | 27 | 44 |
| Bioreactor & scale-up | 5 | 1 | 0 | 17 | 23 |
| Techno-Economic & LCA | 3 | 2 | 6 | 6 | 17 |
| Cell Lines & Engineering | 15 | 1 | 3 | 68 | 87 |
| Single-cell atlases | 1 | 0 | 1 | 41 | 43 |
| Cell-line engineering | 11 | 1 | 2 | 24 | 38 |
| Food Safety | 2 | 1 | 1 | 14 | 18 |
| Allergenicity | 2 | 1 | 1 | 14 | 18 |
| Media & Growth Factors | 13 | 1 | 2 | 22 | 38 |
| Media optimization | 13 | 1 | 1 | 6 | 21 |
| Growth factors | 0 | 0 | 1 | 8 | 9 |
| Serum-free media | 0 | 0 | 0 | 1 | 1 |
| Cryopreservation | 0 | 0 | 0 | 0 | 0 |
| Metabolism & Modeling | 17 | 7 | 3 | 29 | 56 |
| Metabolic modeling | 17 | 7 | 3 | 29 | 56 |
| Scaffolding & Biomaterials | 0 | 0 | 0 | 5 | 5 |
| Scaffolds & biomaterials | 0 | 0 | 0 | 4 | 4 |
| Sensory & Flavor | 12 | 5 | 7 | 39 | 63 |
| Mass spectrometry & metabolomics | 8 | 4 | 3 | 25 | 40 |
| Flavor & sensory prediction | 4 | 1 | 4 | 15 | 24 |
Citation weight
583 entries carry an OpenAlex citation count (331 of 346 papers, 252 of 354 tools, databases and dataset entries); 52 of them sum several release papers into one figure. Bars are shares of the 583 counted, not of the library. An entry with no count is simply not indexed, which is not the same as being uncited. Browse by band on Citations. A popularity signal, not a measure of quality.
Citation weight by subject
The same 583 counted entries, papers included, split by subject. An item tagged with several themes appears under each of them, so rows overlap and add up to more than 583. Any cell opens the resources behind it.
| Subject | 1,000+ | 100–999 | 10–99 | Under 10 | All |
|---|---|---|---|---|---|
| AI Methods & Tooling | 39 | 41 | 51 | 50 | 181 |
| AI agents & foundation models | 4 | 15 | 22 | 28 | 69 |
| Benchmarks & evaluation | 0 | 2 | 9 | 8 | 19 |
| Comparative studies | 0 | 0 | 3 | 1 | 4 |
| Bioprocess & Manufacturing | 2 | 19 | 39 | 34 | 94 |
| Bioreactor & scale-up | 1 | 9 | 22 | 22 | 54 |
| Techno-Economic & LCA | 1 | 8 | 9 | 3 | 21 |
| Cell Lines & Engineering | 17 | 36 | 30 | 26 | 109 |
| Single-cell atlases | 8 | 20 | 15 | 11 | 54 |
| Cell-line engineering | 9 | 12 | 10 | 5 | 36 |
| Food Safety | 2 | 9 | 3 | 4 | 18 |
| Allergenicity | 2 | 9 | 3 | 4 | 18 |
| Media & Growth Factors | 6 | 8 | 26 | 15 | 55 |
| Media optimization | 6 | 8 | 10 | 2 | 26 |
| Growth factors | 0 | 0 | 4 | 1 | 5 |
| Serum-free media | 0 | 0 | 3 | 2 | 5 |
| Cryopreservation | 0 | 0 | 1 | 2 | 3 |
| Metabolism & Modeling | 16 | 29 | 15 | 20 | 80 |
| Metabolic modeling | 16 | 29 | 15 | 20 | 80 |
| Scaffolding & Biomaterials | 0 | 2 | 9 | 7 | 18 |
| Scaffolds & biomaterials | 0 | 2 | 8 | 4 | 14 |
| Sensory & Flavor | 17 | 26 | 61 | 18 | 122 |
| Flavor & sensory prediction | 1 | 15 | 44 | 5 | 65 |
| Mass spectrometry & metabolomics | 17 | 11 | 16 | 9 | 53 |
Momentum
Linked external resources are independent of TUCCA and Tufts University and remain under their own licenses.