Published April 22, 2026
| Version 2026-04-22
Dataset
Open
NPClassifier annotations for the PubChem CID-SMILES dump
Description
NPClassifier predictions for the full PubChem CID-SMILES dump produced offline with recovered NPClassifier weights and a pinned legacy RDKit. This compact release contains row metadata, pathway vectors, superclass vectors, aggregate counts, and sparse class scores aligned to the emitted class ids.
Notes
Dense class vectors are intentionally omitted from the Zenodo package; use the Internet Archive release for the full class matrix.
Files
class-counts.csv
Files
(18.4 GB)
| Name | Size | |
|---|---|---|
|
md5:756cced59452d356e58d7562e717e40d
|
22.6 kB | Preview Download |
|
md5:f8b1efb06f67851a34b200649488e982
|
62.6 kB | Preview Download |
|
md5:c293d27bb91b612d5e6a3f3367d2c845
|
100.2 MB | Download |
|
md5:ad8d90989ade636cf788931ab32c8ad1
|
652 Bytes | Preview Download |
|
md5:52355d995b8e6c8c5f5b65e07de30bdf
|
1.1 kB | Download |
|
md5:a1ff4630fdc05db808d21fe0eb239744
|
224 Bytes | Preview Download |
|
md5:a9c242d01256e27d3a4254e806c2f848
|
1.5 GB | Download |
|
md5:ac2db59f4be553d801b3802bd01f5ee1
|
6.8 kB | Preview Download |
|
md5:4795cd964329d26911096ba3bbdaf40f
|
2.2 GB | Download |
|
md5:e9a5704bfa29c78ab959efed647e5f2b
|
1.0 kB | Preview Download |
|
md5:fe8bb8c2ea5e903f9d52a1e8541301ff
|
2.1 kB | Preview Download |
|
md5:e14fb3ed0cf1e807f02b746629e93447
|
14.7 GB | Download |
|
md5:1c084298d77a8dabb584217e90f8dc75
|
23.0 kB | Preview Download |
Additional details
Related works
- Is supplement to
- Software: https://github.com/earth-metabolome-initiative/npc-labeler (URL)
- References
- Software: https://github.com/mwang87/NP-Classifier (URL)