遇见数据集

Compositional and interpretable representation of histology using AI foundation models and sparse autoencoders

收藏
Zenodo2026-07-31 更新2026-08-02 收录
官方服务:

资源简介:

This repository hosts an index of data and instructions to access original data associated with "Compositional and interpretable representation of histology using AI foundation models andsparse autoencoders". TABLE OF CONTENTS GENERAL INFORMATION FILE ORGANIZATION AWS DATA ACCESS OTHER REPOSITORY FILE LIST ADDITIONAL NOTES/COMMENTS GENERAL INFORMATION Compositional and interpretable representation of histology using AI foundation models and sparse autoencoders Authors: Ziyuan Zhao*, Zoltan Maliga*, Emmanuel C. Ogbonna, Soheil R. Talemi, Shannon Coy, Andréanne Gagné, Kapongo Lumamba, Isaac H. Solomon, Sandro Santagata, Adrie J.C. Steyn, Threnesan Naidoo†, Peter K. Sorger† Please cite this data as the following:Zhao, Z. et al. (2026). Compositional and interpretable representation of histology using AI foundation models and sparse autoencoders. FILE ORGANIZATION Each file follows the following naming convention: Each folder corresponds to a patient sample (N). File Type Description Location N.ome.tif Stitched multiplex CyCIF image pyramid in ome.tif format AWS N.ome.tif Hematoxylin and Eosin stained image of adjacent FFPE tissue section AWS This dataset has multiple parts corresponding to Supplementary Table 1 of the paper: Dataset name Disease type Number of H&E WSIs Medical center DS1 Pulmonary TB 40 Brigham and Women's Hospital DS2 Pulmonary TB 14 Brigham and Women's Hospital DS3 Pulmonary TB 20 University of KwaZulu-Natal (AHRI) DS4 Lung adenocarcinoma 12 Brigham and Women's Hospital DS5 Sarcoidosis 5 Stockholm University and Research Center Borstel AWS Data Access DS1-5 will be available from AWS. For DS1, you will need the following bucket name: lsp-public-data/ogbonna-2026-tb-tls/ For DS2-5, you will need the following bucket name: lsp-public-data/zhao-2026-fmsae/ For instructions on how to access data from AWS, visit “Access Laboratory of Systems Pharmacology Datasets on AWS” (https://zenodo.org/records/10223574) If you experience issues accessing the above AWS S3 buckets, email tissue-atlas(at)hms.harvard.edu with the subject line "ogbonna-2026-tb-tls: Data Access" or "zhao-2026-fmsae: Data Access". Other Repository DS5 is also available from the original paper's figshare page (https://figshare.com/articles/figure/_b_Immune_mapping_of_human_tuberculosis_and_sarcoidosis_lung_granulomas_b_/24793848/1) Please download S1 HE.tif, S2 HE.tif, S4 HE.tif, S5 HE.tif, S6 HE.tif from the figshare page. Our version of DS5 pyramidalized the whole slide images for efficient viewing and processing. FILE LIST HE Image Data DS1: File Name Location File Size (GB) LSP24521.ome.tif AWS 12.15 LSP24524.ome.tif AWS 14.00 LSP24527.ome.tif AWS 13.48 LSP24530.ome.tif AWS 12.78 LSP24533.ome.tif AWS 13.78 LSP24536.ome.tif AWS 16.61 LSP24539.ome.tif AWS 16.72 LSP24542.ome.tif AWS 16.42 LSP24545.ome.tif AWS 18.33 LSP24548.ome.tif AWS 13.97 LSP24551.ome.tif AWS 12.54 LSP24554.ome.tif AWS 13.91 LSP24557.ome.tif AWS 13.83 LSP24560.ome.tif AWS 13.21 LSP24563.ome.tif AWS 13.25 LSP24566.ome.tif AWS 13.93 LSP24569.ome.tif AWS 13.46 LSP24572.ome.tif AWS 14.41 LSP24575.ome.tif AWS 13.23 LSP24578.ome.tif AWS 13.43 LSP24581.ome.tif AWS 13.23 LSP24584.ome.tif AWS 12.64 LSP24587.ome.tif AWS 13.73 LSP24590.ome.tif AWS 13.01 LSP24593.ome.tif AWS 12.73 LSP24596.ome.tif AWS 13.72 LSP24599.ome.tif AWS 17.65 LSP24602.ome.tif AWS 11.89 LSP24605.ome.tif AWS 16.95 LSP24608.ome.tif AWS 14.84 LSP24611.ome.tif AWS 13.68 LSP24614.ome.tif AWS 13.78 LSP24617.ome.tif AWS 13.49 LSP24620.ome.tif AWS 14.22 LSP24623.ome.tif AWS 13.12 LSP24626.ome.tif AWS 13.01 LSP24629.ome.tif AWS 13.53 LSP24632.ome.tif AWS 12.78 LSP24635.ome.tif AWS 13.39 LSP24638.ome.tif AWS 11.72 DS2: File Name Location File Size LSP11103.ome.tif AWS 957M LSP16106.ome.tif AWS 942M LSP18955.ome.tif AWS 473M LSP18958.ome.tif AWS 518M LSP18967.ome.tif AWS 649M LSP18970.ome.tif AWS 915M LSP18973.ome.tif AWS 759M LSP18976.ome.tif AWS 709M R21-B5.ome.tif AWS 749M SANDRO6slide.ome.tif AWS 493M Z116-TB_HE.ome.tif AWS 751M DS3: File Name Location File Size SL238-AH1.ome.tif AWS 5.8G SL238-AM1.ome.tif AWS 4.0G SL412-A3.ome.tif AWS 6.0G SL464-C2.ome.tif AWS 2.1G SL470-A2.ome.tif AWS 3.0G SL470-A8.ome.tif AWS 1.9G SL470-C1.ome.tif AWS 3.2G SL479-B1.ome.tif AWS 4.6G SL495-B1.ome.tif AWS 3.5G SL574-A3.ome.tif AWS 2.5G SL579-A1.ome.tif AWS 1.4G SL628-B3.ome.tif AWS 2.7G SL717-A1.ome.tif AWS 2.3G SL717-A6.ome.tif AWS 3.0G SL724-B3.ome.tif AWS 2.2G SL750-B1.ome.tif AWS 5.5G SL752-A2.ome.tif AWS 2.3G SL814-A4.ome.tif AWS 2.5G SL814-B2.ome.tif AWS 4.0G SL814-B5.ome.tif AWS 2.4G DS4: File Name Location File Size P132115_HE-0.ome.tif AWS 5.2G P132630_HE-0.ome.tif AWS 7.3G P132666_HE-0.ome.tif AWS 8.8G P133537_HE-0.ome.tif AWS 5.5G P136690_HE-0.ome.tif AWS 4.4G P137591_HE-0.ome.tif AWS 6.1G P137753_HE-0.ome.tif AWS 8.2G P137757_HE-0.ome.tif AWS 8.8G P137905_HE-0.ome.tif AWS 7.4G P137941_HE-0.ome.tif AWS 9.6G P137974_HE-0.ome.tif AWS 8.9G P138007_HE-0.ome.tif AWS 11G DS5: File Name Location File Size (GB) S1_HE.ome.tif AWS 161M S2_HE.ome.tif AWS 438M S4_HE.ome.tif AWS 287M S5_HE.ome.tif AWS 280M S6_HE.ome.tif AWS 284M CyCIF Image Data DS1: File Name Location File Size (GB) LSP24522.ome.tif AWS 222.37 LSP24525.ome.tif AWS 202.51 LSP24528.ome.tif AWS 202.57 LSP24531.ome.tif AWS 225.94 LSP24534.ome.tif AWS 220.79 LSP24537.ome.tif AWS 204.92 LSP24540.ome.tif AWS 191.57 LSP24543.ome.tif AWS 213.09 LSP24546.ome.tif AWS 205.37 LSP24549.ome.tif AWS 212.91 LSP24552.ome.tif AWS 199.53 LSP24555.ome.tif AWS 212.91 LSP24558.ome.tif AWS 221.39 LSP24561.ome.tif AWS 217.61 LSP24564.ome.tif AWS 214.29 LSP24567.ome.tif AWS 196.63 LSP24570.ome.tif AWS 207.13 LSP24573.ome.tif AWS 226.64 LSP24576.ome.tif AWS 230.50 LSP24579.ome.tif AWS 205.65 LSP24582.ome.tif AWS 226.38 LSP24585.ome.tif AWS 199.04 LSP24588.ome.tif AWS 215.38 LSP24591.ome.tif AWS 200.37 LSP24594.ome.tif AWS 217.11 LSP24597.ome.tif AWS 199.92 LSP24600.ome.tif AWS 202.57 LSP24603.ome.tif AWS 219.74 LSP24606.ome.tif AWS 219.70 LSP24609.ome.tif AWS 226.05 LSP24612.ome.tif AWS 197.44 LSP24615.ome.tif AWS 222.30 LSP24618.ome.tif AWS 197.84 LSP24621.ome.tif AWS 198.85 LSP24624.ome.tif AWS 198.49 LSP24627.ome.tif AWS 216.52 LSP24630.ome.tif AWS 210.79 LSP24633.ome.tif AWS 193.94 LSP24636.ome.tif AWS 203.49 LSP24639.ome.tif AWS 185.99 ADDITIONAL NOTES/COMMENTS Please contact tissue-atlas(at)hms.harvard.edu if you have any questions related to the data.

提供机构:
Zenodo
创建时间:
2026-07-12
二维码
社区交流群
二维码
科研交流群
商业服务