{"success":true,"data":{"author":"Thomas, Belanger, Xu et al. (Google DeepMind and Google Research, with collaborators)","bucketAccessType":"public","bucketListable":true,"bucketName":"nuclease_design","bucketPrefix":"","category":"Proteomics","cloudProvider":"gcp","createdAt":"2026-11-19T15:09:40.262Z","description":"Data from Thomas, Belanger et al., 'Engineering highly active and diverse nuclease enzymes by combining machine learning and ultra-high-throughput screening' (bioRxiv 2024), in which ML models (including a VAE and a CNN) proposed NucB nuclease variants that were screened in four rounds of ultra-high-throughput sorting. The bucket holds raw_count_data/ (NGS counts for pre-sort and post-sort populations, rounds G1 to G4), processed_fiducial_data/ (enrichment factors for synonymous DNA sequences of fiducial variants, used to build null distributions), processed_data/ (per amino acid sequence enrichment factors and p-values for g1.csv to g4.csv, and landscape.csv, which merges all rounds into multi-class catalytic activity labels for about 56,000 distinct sequences; the README describes the estimated activity of 55,760 variants), plate_data/ (low-throughput purified protein assays confirming hits), library_designs/ (which design method proposed each sequence), alignments/ (the multiple sequence alignment used for the VAE), cnn_models/ and analysis/. Load CSVs with pandas; the nuclease_design package and Colab notebooks reproduce the figures. DeepMind licenses all non-software materials CC BY 4.0; cite Thomas, Belanger et al. 2024.","displayTitle":"ML-Guided Directed Evolution of the NucB Nuclease: 55,760-Variant Activity Landscape and NGS Counts (Google DeepMind)","doi":"10.1101/2024.03.21.585615","fuseMountBrowsable":true,"id":"97111039-569b-5e03-8084-033a5abc8b74","license":"CC-BY-4.0","licenseEvidenceUrl":"https://github.com/google-deepmind/nuclease_design","objectCountEstimate":8,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1101/2024.03.21.585615","region":"us","schemaCard":{"accessHints":[{"language":"python-gcsfs","snippet":"import gcsfs\nfs = gcsfs.GCSFileSystem(token='anon')\nfs.ls('nuclease_design/', detail=False)[:10]"},{"language":"python-storage","snippet":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nlist(client.bucket('nuclease_design').list_blobs(prefix='', max_results=10))"}],"caveats":["Files are binary — use a domain-specific parser rather than text I/O."],"format":"binary","generatedAt":"2026-11-19T15:22:22.207Z","layout":{"example":"alignments/F1BV52_BACLI_e3_n8_mFirst.a2m","pattern":"{prefix}/{filename}"},"mountPath":"/mnt/gcs-archive/data","sampleHeaders":[{"bytes":">tr|F1BV52|F1BV52_BACLI NucB OS=Bacillus licheniformis OX=1402 GN=nucB PE=1 SV=1\nMIKKW.......AVHLLFSALV..............LLGL.S...........G....GA...A....YS.........P.....Q.H..A.........\n..E...G......A...A....R....................Y.....DD.....V..LYF..P...........A........S..............\n...R.......Y...P.......E...T.........G..A.HIS.D...A....I...........K................................\n.A............................G...............................H.....................................\n..........A................","encoding":"utf8","key":"alignments/F1BV52_BACLI_e3_n8_mFirst.a2m","truncatedAt":512}],"sampleObjects":["alignments/F1BV52_BACLI_e3_n8_mFirst.a2m"],"sidecars":[{"bytes":"# ML-Guided Directed Evolution for Engineering a Better Nuclease Enzyme\n\n\n## Data\n\nData is available in a Google cloud storage bucket:\n`https://storage.googleapis.com/nuclease-design-data`.\n\nIt is organized in the following sub-directories:\n\n*   `raw_count_data`: raw NGS count data for pre-sort and post-sort populations.\n\n*   `processed_fiducial_data`: enrichment factors for synonyms of various\n    'fiducial' sequences. Each row represents a distinct DNA sequence that\n    translates to the same amino acid sequence.\n\n*   `processed_data`: enrichment factors computed from the raw count data and\n    the processed fiducial data. Each row represents a unique amino acid\n    sequence. For each row and each fiducial, the row is assigned a p-value for\n    observing its enrichment factor under the null distribution of enrichment\n    factors from the fiducial.\n\n*   `processed_data/landscape.csv`: A single file that merges data from all 4\n    rounds of experiments and provides a multi-class catalytic activity\n    labels for 56K distinct amino acid sequences.\n\n*   `plate_data`: Data from the low-throughput purified protein experiments used\n    to confirm hits.\n\n*   `library_designs`: A mapping from amino acid sequences to the list of the\n    names of the sub-libraries (corresponding to different sequence design\n    methods) that proposed it. Note that some sequences were proposed by\n    multiple methods.\n\n*   `analysis`: Data used for creating certain tables and results in the paper\n    that require expensive computations, such as clustering hits in order to\n    quantify diversity.\n\n*   `alignments`: A multiple sequence alignment used to fit our VAE model.\n","encoding":"utf8","key":"README.md","name":"README.md","size":1672,"truncated":false}],"sourceObjectKey":"alignments/F1BV52_BACLI_e3_n8_mFirst.a2m","starterCell":"from google.cloud import storage\nclient = storage.Client.create_anonymous_client()\nbucket = client.bucket('nuclease_design')\nblob = bucket.blob('alignments/F1BV52_BACLI_e3_n8_mFirst.a2m')\nwith blob.open(\"rb\") as fh:\n    print(fh.read(64))","starterSamplePath":"alignments/F1BV52_BACLI_e3_n8_mFirst.a2m","totalObjects":"8 top-level prefixes"},"summary":"Data from Thomas, Belanger et al., 'Engineering highly active and diverse nuclease enzymes by combining machine learning and ultra-high-throughput screening' (bioRxiv 2024), in…","title":"ML-Guided Directed Evolution of the NucB Nuclease: 55,760-Variant Activity Landscape and NGS Counts (Google DeepMind)","updatedAt":"2026-11-19T19:03:20.352Z","visibility":"public","websiteUrl":"https://github.com/google-deepmind/nuclease_design"}}