{"success":true,"data":{"author":"DeepMind and Calico Life Sciences","bucketAccessType":"public","bucketListable":true,"bucketName":"dm-enformer","bucketPrefix":"","category":"Genomics & Bioinformatics","cloudProvider":"gcp","createdAt":"2026-11-14T01:00:54.940Z","description":"Public release accompanying Enformer (Avsec et al., Nature Methods 2021), DeepMind and Calico's transformer model that predicts gene expression and chromatin tracks from up to about 100 kb of DNA sequence context by integrating long-range interactions. Measured 2026-11-13, the bucket has three top-level folders. models/enformer/ holds the TensorFlow SavedModel (saved_model.pb, variables/) and Sonnet weights, plus fitted robust-scaler and PCA transforms for SAD (SNP activity difference) scores. data/gtex_fine/ holds GTEx fine-mapped eQTL sets per tissue: SuSiE output tables (.tsv, tens of MB per tissue) and positive/negative variant VCFs per tissue used to evaluate variant effect prediction. variant-scores/1000-genomes/enformer/ holds precomputed Enformer variant effect predictions for 1000 Genomes variants with minor allele frequency above 0.005, as one HDF5 file per chromosome (individual files range up to about 42 GB), plus a PC20 subfolder. Open the HDF5 files with h5py, the TSVs with pandas, the VCFs with pysam or cyvcf2, and the model with tensorflow / tensorflow_hub. Use cases: variant effect prediction, eQTL prioritization, regulatory genomics and sequence-to-expression deep learning. Copyright 2021 DeepMind Technologies, released under CC BY 4.0 per the in-bucket README; cite the Enformer paper.","displayTitle":"Enformer Model, GTEx Fine-Mapped eQTLs and 1000 Genomes Variant Effect Scores","doi":"10.1038/s41592-021-01252-x","fuseMountBrowsable":true,"id":"0577c367-38fb-5c8b-bbd9-0b10b4291dbc","license":"CC-BY-4.0","licenseEvidenceUrl":"https://storage.googleapis.com/dm-enformer/README","objectCountEstimate":3,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1038/s41592-021-01252-x","region":"us","schemaCard":{"accessHints":[{"language":"python-gcsfs","snippet":"import gcsfs\nfs = gcsfs.GCSFileSystem(token='anon')\nfs.ls('dm-enformer/', detail=False)[:10]"},{"language":"python-storage","snippet":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nlist(client.bucket('dm-enformer').list_blobs(prefix='', max_results=10))"}],"caveats":["Files are binary — use a domain-specific parser rather than text I/O."],"format":"binary","generatedAt":"2026-11-14T01:01:04.358Z","layout":{"example":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","pattern":"{id}/{seg1}/{seg2}/{filename}"},"mountPath":"/mnt/gcs-archive/data","sampleHeaders":[{"bytes":"\tchromosome\tstart\tend\tvariant\tallele1\tallele2\tminor_allele\ttissue\tgene\tmaf\tbeta_marginal\tse_marginal\tz\tpip\tcs_id\tbeta_posterior\tsd_posterior\tsnp\tvariant_hg19\n0\tchr1\t13549\t13550\tchr1_13550_G_A_b38\tG\tA\tA\tAdipose_Subcutaneous\tENSG00000227232.5\t0.0154905\t0.24370100000000003\t0.201673\t1.2083967610934498\t0.0074531088394888\t-1\t0.000569217642800581\t0.0112152474763868\tTrue\tchr1_13550_G_A\n1\tchr1\t14670\t14671\tchr1_14671_G_C_b38\tG\tC\tC\tAdipose_Subcutaneous\tENSG00000227232.5\t0.010327\t0.41984399999999994\t0.24771500000000002","encoding":"utf8","key":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","truncatedAt":512},{"bytes":"\tchromosome\tstart\tend\tvariant\tallele1\tallele2\tminor_allele\ttissue\tgene\tmaf\tbeta_marginal\tse_marginal\tz\tpip\tcs_id\tbeta_posterior\tsd_posterior\tsnp\tvariant_hg19\n0\tchr1\t14676\t14677\tchr1_14677_G_A_b38\tG\tA\tA\tAdipose_Visceral_Omentum\tENSG00000228327.3\t0.0522388\t0.889035\t0.11251300000000002\t7.90162025721472\t0.9999999993375792\t1\t0.8208189390013829\t0.103463910499492\tTrue\tchr1_14677_G_A\n1\tchr1\t14676\t14677\tchr1_14677_G_A_b38\tG\tA\tA\tAdipose_Visceral_Omentum\tENSG00000228463.9\t0.0522388\t-0.901493\t0.120957\t-7.45300396008499","encoding":"utf8","key":"data/gtex_fine/susie/Adipose_Visceral_Omentum.tsv","truncatedAt":512}],"sampleObjects":["data/gtex_fine/susie/Adipose_Subcutaneous.tsv","data/gtex_fine/susie/Adipose_Visceral_Omentum.tsv","data/gtex_fine/susie/Adrenal_Gland.tsv","data/gtex_fine/susie/Artery_Aorta.tsv","data/gtex_fine/susie/Artery_Coronary.tsv"],"sidecars":[{"bytes":"Copyright (2021) DeepMind Technologies Limited, made available under the Creative Commons Attribution 4.0 License. A full copy of the license can be found at https://creativecommons.org/licenses/by/4.0/.\n\nIf you use the dataset you should cite the associated paper:\n\n\"Effective gene expression prediction from sequence by integrating long-range\ninteractions\"\n\nŽiga Avsec1, Vikram Agarwal2,4, Daniel Visentin1,4, Joseph R. Ledsam1,3,\nAgnieszka Grabska-Barwinska1, Kyle R. Taylor1, Yannis Assael1, John Jumper1,\nPushmeet Kohli1, David R. Kelley2*\n\n1 DeepMind, London, UK\n2 Calico Life Sciences, South San Francisco, CA, USA\n3 Google, Tokyo, Japan\n4 These authors contributed equally.\n* correspondence: avsec@google.com, pushmeet@google.com, drk@calicolabs.com\n","encoding":"utf8","key":"README","name":"README","size":759,"truncated":false}],"sourceObjectKey":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","starterCell":"import os\nimport pandas as pd\ndf = pd.read_csv(os.path.join(os.environ['SCIGANTIC_MOUNT_PATH'], 'data/gtex_fine/susie/Adipose_Subcutaneous.tsv'), nrows=1000)\ndf.head()","starterSamplePath":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","totalObjects":"3 top-level prefixes","verifiedRead":{"attempts":[{"family":"csv","ok":true,"path":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","seconds":1.33}],"bytes":112519610,"facts":[],"family":"csv","image":"531775725781.dkr.ecr.us-east-1.amazonaws.com/scigantic-notebook-genomics:latest","library":"pandas","path":"data/gtex_fine/susie/Adipose_Subcutaneous.tsv","replacedStarterCell":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nbucket = client.bucket('dm-enformer')\nblob = bucket.blob('data/gtex_fine/susie/Adipose_Subcutaneous.tsv')\nwith blob.open(\"rb\") as fh:\n    print(fh.read(64))","rssDeltaMb":72.7,"seconds":1.33,"snippet":"import os\nimport pandas as pd\ndf = pd.read_csv(os.path.join(os.environ['SCIGANTIC_MOUNT_PATH'], 'data/gtex_fine/susie/Adipose_Subcutaneous.tsv'), nrows=1000)\ndf.head()","starterCell":{"error":"ModuleNotFoundError: No module named 'google.cloud'","ok":false,"seconds":0},"summary":"CSV/TSV: columns ['Unnamed: 0', 'chromosome', 'start', 'end', 'variant', 'allele1', 'allele2', 'minor_allele', 'tissue', 'gene', 'maf', 'beta_marginal'], dtypes ['int64', 'object', 'int64', 'int64', 'object', 'object', 'object', 'object', 'object', 'object', 'float64', 'float64'], first 1000 rows read","verifiedAt":"2026-11-14T06:15:59.060Z"}},"summary":"Public release accompanying Enformer (Avsec et al., Nature Methods 2021), DeepMind and Calico's transformer model that predicts gene expression and chromatin tracks from up to…","title":"Enformer Model, GTEx Fine-Mapped eQTLs and 1000 Genomes Variant Effect Scores","updatedAt":"2026-11-14T06:15:59.060Z","visibility":"public","websiteUrl":"https://github.com/google-deepmind/deepmind-research/tree/master/enformer"}}