Data files
Release 2026.11, created 2026-11-10. Tables are comma-separated, UTF-8, and gzip-compressed; the correlation matrix is HDF5. Expand a row for its columns and SHA-256 checksum.
| Dataset | Format | Size | Rows | Download |
|---|---|---|---|---|
|
Predicted functions (term rankings)
The 50 highest and 50 lowest mean-correlation terms for every snoRNA in each Enrichr library shown on the website (ChEA 2022 is used only for the similarity map and is not included).
Columns and checksum
SHA-256:
07caf3a03669e8bc47681f909525807327162587d59931505307c04a58781297
|
CSV, gzip-compressed, UTF-8 | 107.3 MB | 494,000 | top_terms_snoRNA.csv.gz |
|
Genes behind each predicted term
For every ranked term, the term genes with the highest correlation to the snoRNA (up to 25 per term).
Columns and checksum
SHA-256:
29b3c0a348d60c17697f7befe02f6585fc7d3d16d9a08dee588a649820b6e48e
|
CSV, gzip-compressed, UTF-8 | 159.7 MB | 6,569,350 | ranked_overlap_genes.csv.gz |
|
Top correlated protein-coding genes
The 500 most positively and 500 most negatively correlated protein-coding genes for every snoRNA.
Columns and checksum
SHA-256:
29015f3bdb8c9cd4410f8f6405ffa931768dbcc44ee0861f8bbdd755684c30ac
|
CSV, gzip-compressed, UTF-8 | 22.8 MB | 988,000 | top_correlated_genes_snoRNA.csv.gz |
|
Upstream transcription factors (ChEA3)
The top 25 transcription factors ChEA3 ranks for each snoRNA's top 500 positively correlated genes, by integrated mean rank.
Columns and checksum
SHA-256:
eed45e96f03c1d78a69a84e5e51164d77567661c4d3195d568451581b536adc1
|
CSV, gzip-compressed, UTF-8 | 2.4 MB | 24,700 | snoRNA_chea3_transcription_factors.csv.gz |
|
Perturb-Seqr enrichment results
Drug and gene perturbation signatures enriched among each snoRNA's top 500 positively correlated genes.
Columns and checksum
SHA-256:
9a28ab9c3fe3ce29b913331b665ac9939402fabceb453287e0b84fda236fb745
|
CSV, gzip-compressed, UTF-8 | 11.0 MB | 131,887 | all_snornas_perturbseqr_results.csv.gz |
|
Perturb-Seqr results across libraries
The website's drug and gene perturbation tables for every snoRNA, ranked by the number of supporting libraries, with each library's rank and adjusted p-value and any conflicting libraries.
Columns and checksum
SHA-256:
4d00871ecd186c5f32d872ac808a84964f51402b1ef266f56c7bfd0ba6c17b8b
|
CSV, gzip-compressed, UTF-8 | 14.9 MB | 156,571 | snoRNA_perturbseqr_consensus.csv.gz |
|
snoRNA similarity map
UMAP and t-SNE coordinates and the ten nearest snoRNAs by cosine similarity of term profiles.
Columns and checksum
SHA-256:
a99ed60befff8bfaadc7e76f9b559620dd267043d65aa6367a962a76b8edd488
|
CSV, gzip-compressed, UTF-8 | 73.6 KB | 988 | snoRNA_similarity.csv.gz |
|
snoRNA sequences and predicted structures
RNA sequence, dot-bracket secondary structure, and the method and template used for every snoRNA.
Columns and checksum
SHA-256:
c4498a94d404e1c1ecac80c98f21e08df6f2ae4a9f8987b62cfd6c67fc13f763
|
CSV, gzip-compressed, UTF-8 | 64.2 KB | 988 | snoRNA_structures.csv.gz |
|
Protein-coding gene filter
Keep or drop decision, with reason, for every ARCHS4 protein-coding row used in the rankings.
Columns and checksum
SHA-256:
22126c470eecc92b60ba9297261670c6e8df3f444e16f7874cc6dff3c7490e10
|
CSV, gzip-compressed, UTF-8 | 363.9 KB | 22,805 | protein_coding_gene_filter.csv.gz |
|
Full correlation matrix (HDF5)
Pearson r, p-value and number of shared samples for every protein-coding gene and snoRNA pair, before the gene filter.
Datasets and checksum
SHA-256:
07051646f979b25aec225652782c15699c08ff5d793c1e6417e68f48099522be
|
HDF5 | 198.8 MB | correlation.h5 |
Verify a file after download with
shasum -a 256 <file>
(macOS) or
sha256sum <file>
(Linux).
Programmatic access
The data behind each report page is available as JSON. No key or registration is needed.
| Endpoint | Returns |
|---|---|
GET /api/snornas |
Every snoRNA symbol and Ensembl ID that can be queried |
GET /api/snorna/<symbol or Ensembl ID> |
The full report: top positively and negatively correlated genes, predicted terms by library, and Perturb-Seqr results. A symbol whose Ensembl records have different results returns status 300 with those records in matches; request one by its ensembl_id for its report. |
GET /api/snorna/<query>/annotation |
Gene annotation for the snoRNA |
GET /api/snorna/<query>/location |
Genomic location of the snoRNA |
GET /api/similarity |
The snoRNA similarity map |
POST /api/set-analysis |
Set analysis for a JSON body such as {"snornas": ["SNORD116-1", "SNORD116-2"]}. Add "direction": "negative" for the negatively correlated genes; the default is "positive". The Perturb-Seqr and ChEA3 services take the same body. |
POST /api/set-analysis/perturbseqr |
Perturb-Seqr results for a set, with the same JSON body |
POST /api/set-analysis/chea3 |
ChEA3 results for a set, with the same JSON body |
Example:
curl https://snohub2.k8s.maayanlab.cloud/api/snorna/SNORD116-1
Terms of use
snoHUB2 and all files on this page are free to use, with no login or registration.
The predictions are derived from public resources that keep their own terms. Gene-set term names and memberships come from the Enrichr libraries listed in the Methods, and some of those libraries carry their own licence conditions. Predictions are based on coexpression and do not establish causality.
Data sources
- ARCHS4: human bulk RNA-seq gene counts
- Enrichr: gene-set libraries used for function prediction
- Perturb-Seqr (preprint): drug and gene perturbation signatures
- Ensembl release 107 and HGNC: gene biotypes and symbols for the protein-coding filter
- Ensembl REST API and RNAcentral: snoRNA sequences and R2DT secondary structures
The full workflow is described on the Methods page.