De-identified patient-level cell-composition data for SMMG-AML
Authors/Creators
Description
This dataset contains de-identified patient-level cell-type composition histograms and subtype-level composition summaries for the retrospective two-center cohort of 199 participants analyzed in SMMG-AML (manuscript ID BSPC-D-25-14932). Each patient-level record contains a randomized participant identifier, patient-level five-fold cross-validation assignment, molecularly confirmed diagnostic group, retained single-cell crop count, and 21 model-estimated cell-type fractions.
The fractions are raw predicted-label proportions before the Laplacian smoothing used for downstream fusion and are not locally validated cell-level ground-truth labels.The release contains no raw or cropped bone-marrow smear images, single-cell feature embeddings, patient-level model prediction probabilities, source code, executable analysis scripts, model weights, or the private accession-to-anonymized-ID mapping. Raw images may be requested by qualified academic researchers for non-commercial research under institutional and ethics approval and a data-use agreement.
Funding: This work was supported by the Guangdong Basic and Applied Basic Research Foundation (2024A1515010002), the Guizhou Province Science and Technology Support Program (Qian Science and Technology Support [2024] General Project 108), and the National Natural Science Foundation of China (61975069).
Files
data_dictionary.csv
Files
(87.8 kB)
| Name | Size | |
|---|---|---|
|
md5:7961878794e0e4f80e3fc651601e69a7
|
7.3 kB | Preview Download |
|
md5:b2f5215d48928b8d489ef65a6ba53101
|
3.6 kB | Preview Download |
|
md5:208a6359a388e272abd9d8293af849ca
|
471 Bytes | Preview Download |
|
md5:55fab0d01f7cc8721ff7f031851f6755
|
62.3 kB | Preview Download |
|
md5:75a038de65eccbf8be50623c2f278a44
|
3.4 kB | Preview Download |
|
md5:993afcd7e1251cfc2dc1dd8a9de2dced
|
536 Bytes | Preview Download |
|
md5:b37ba5fe7610cb28f7218fa5bd1ccbd3
|
10.2 kB | Preview Download |
Additional details
Funding
- National Natural Science Foundation of China
- 61975069