{"success":true,"data":{"author":"Ferebee, Wrightsman, Zhai, Gokaslan, Kuleshov and Buckler (Cornell University)","bucketAccessType":"public","bucketName":"maize-genetics/plexbench-base","bucketPrefix":"","category":"Genomics & Bioinformatics","cloudProvider":"huggingface","createdAt":"2026-11-21T22:15:36.027Z","description":"Benchmark data for predicting plant gene expression from DNA sequence, from PLExBench: A benchmarking suite for predicting gene expression in plants, curated by Taylor Ferebee, Travis Wrightsman, Jingjing Zhai, Aaron Gokaslan, Volodymyr Kuleshov and Edward S. Buckler (Cornell University). Expression comes from public RNA-seq runs in the NCBI Sequence Read Archive for leaf and root tissue of Arabidopsis thaliana Col-0 and maize (Zea mays) B73 seedlings, quantified as transcripts per million (TPM). Three task types are derived per species and tissue: maximum expression across experiments, absolute (mean) expression, and on/off expression. Genes are split 80/10/10 into train, validation and test by orthogroup-guided splitting (Washburn et al. 2019) so related genes do not leak across splits. The repository holds two zip archives (about 0.8 GB): tasks/ with species-task-tissue folders of train, validate and test TSV files, and genomes/ with annotation FASTA and GFF files for each genome. Useful for training and evaluating DNA language models and sequence-to-expression predictors in crops. Released by the authors under the MIT License; cite the PLExBench paper.","displayTitle":"PLExBench: Maize and Arabidopsis Leaf and Root Gene Expression Benchmark for Sequence-to-Expression Models (Buckler Lab, Cornell)","huggingFaceUrl":"https://huggingface.co/datasets/maize-genetics/plexbench-base","id":"444c8c29-9237-540a-ba96-f132a53ef762","license":"MIT","licenseEvidenceUrl":"https://huggingface.co/datasets/maize-genetics/plexbench-base","objectCountEstimate":4,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/maize-genetics/plexbench-base","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"maize-genetics/plexbench-base\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"maize-genetics/plexbench-base\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=maize-genetics, license:mit, downloads=164"],"format":"mixed","generatedAt":"2026-11-21T22:31:49.306Z","layout":{"example":"genomes.zip","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"genomes.zip","truncatedAt":512}],"sampleObjects":["genomes.zip","tasks.zip"],"sidecars":[{"bytes":"---\nlicense: mit\ntags:\n- biology\n- plants\n- gene expression\npretty_name: Maize and Arabidopsis gene expression\n---\n# Dataset Card for Maize and Arabidopsis gene expression\nPlant Gene expression data used for benchmarking sequence to gene expression prediction ML models. \n\n\n### Dataset Description\n\nSpecies included are Maize and Arabidopsis thaliana. Dataset includes gene expression values for leaf and root tissues. \nWithin the `tasks` folder, datasets are broken down by `species-task-tissue`. Genomes in the `genomes` folders include the annotation and the GFF files associated with that specific genome.\nAll tasks are split by 80% train, 10% validation, and 10% test. \n\n## Dataset Structure\n\n```\ndataset\n  genomes/\n    Arabidopsis_thaliana/\n      annotation.fa\n      ath.gff\n    Zea_mays/\n      annotation.fa\n      ath.gff\n  tasks/\n    species-task-tissue/\n      train.tsv\n      validate.tsv\n      test.tsv\n```\n\n- **Curated by:** Taylor Ferebee, Travis Wrightsman, Jingjing Zhai, Aaron Gokaslan, Volodymyr Kuleshov, Edward S. Buckler\n- **Repository:** [https://github.com/maize-genetics/expression-survey]\n- **Paper:** PLExBench: A benchmarking suite for predicting gene expression in plants\n- **License:** MIT\n\n### Dataset Sources \n| | | | | | | | | | | | |\n|-|-|-|-|-|-|-|-|-|-|-|-|\n|sample_name|species|genotype|library_layout|library_selection|reads_location|organ|age|condition|replicate|batch|reference|\n|SRR505743|Arabidopsis_thaliana|Col-0|single-read|random|sra|root|seedling|controlled|1|1|SRP013631|\n|SRR505744|Arabidopsis_thaliana|Col-0|single-read|random|sra|leaf|seedling|controlled|1|1|SRP013631|\n|SRR953400|Arabidopsis_thaliana|Col-0|single-read|random|sra|leaf|seeding|controlled|1|1|PRJNA215448|\n|SRR1005386|Arabidopsis_thaliana|Col-0|single-read|random|sra|leaf|seedling|controlled|1|1|PRJNA222364|\n|SRR578947|Arabidopsis_thaliana|Col-0|single-read|random|sra|root|seedling|controlled|1|1|SRP013631|\n|SRR578948|Arabidopsis_thaliana|Col-0|single-read|random|sra|root|seedling|controlled|1|1|SRP013631|\n|ERR2096663|Zea_mays|B73|paired-end|polyA|sra|leaf|seedling|controlled|1|1|PRJEB22166|\n|ERR2096664|Zea_mays|B73|paired-end|polyA|sra|leaf|seedling|controlled|1|1|PRJEB22166|\n|ERR2096665|Zea_mays|B73|paired-end|polyA|sra|leaf|seedling|controlled|1|1|PRJEB22166|\n|ERR2096666|Zea_mays|B73|paired-end|polyA|sra|leaf|seedling|controlled|1|1|PRJEB22166|\n|ERR2096667|Zea_mays|B73|paired-end|polyA|sra|leaf|seedling|controlled|1|1|PRJEB22166|\n|ERR3773807|Zea_mays|B73|paired-end|polyA|sra|root|seedling|controlled|1|1|PRJEB35943|\n|ERR3773808|Zea_mays|B73|paired-end|polyA|sra|root|seedling|controlled|1|1|PRJEB35943|\n|ERR986091|Zea_mays|B73|paired-end|random|sra|root|seedling|controlled|1|1|PRJEB10406|\n\n\n\n\n\n\n\n\n\n### Curation Rationale\n To choose experiments for leaf and root tissues, we focused on datasets that have been used in a recent study and can be found in\nmultiple databases.\n\n#### Data Collection and Processing\nIn the max gene expression datasets, for each gene, we take the maximum transcript per million TPM value over experiments. Similarly, for the\nabsolute expression datasets, we take the mean TPM value over experiments. Finally, for the on-off ex-\npression, we assign 1 to a gene if it has a TPM value in one of the tissues. To create train-test-validation\nsplits, we use orthogroup guided splitting as introduced by Washburn et al. 2019. Then, we split the training test sets so\nthat we train on 80% of the orthogroups and test on 10%. Note that for each of the task-based datasets, we\nkeep the same train-test-validate split. \n\n**BibTeX:**\n\n\n## Dataset Card Authors\nTaylor Ferebee (tf259@cornell.edu)\n\n## Dataset Card Contact\nTaylor Ferebee (tf259@cornell.edu), Cinta Romay, Edward S. Buckler","encoding":"utf8","key":"README.md","name":"README.md","size":3733,"truncated":false}],"sourceObjectKey":"genomes.zip","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"maize-genetics/plexbench-base\", filename=\"genomes.zip\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"genomes.zip","totalObjects":4},"summary":"Benchmark data for predicting plant gene expression from DNA sequence, from PLExBench: A benchmarking suite for predicting gene expression in plants, curated by Taylor Ferebee,…","title":"PLExBench: Maize and Arabidopsis Leaf and Root Gene Expression Benchmark for Sequence-to-Expression Models (Buckler Lab, Cornell)","updatedAt":"2026-11-22T01:05:08.180Z","visibility":"public","websiteUrl":"https://github.com/maize-genetics/expression-survey"}}