{"success":true,"data":{"author":"Genomes to Fields Initiative","bucketAccessType":"public","bucketListable":true,"bucketName":"scigantic-g2f-maize","bucketPrefix":"","category":"Genomics & Bioinformatics","cloudProvider":"aws","createdAt":"2026-11-16T02:24:18.224Z","description":"Genomes to Fields (G2F) is a public maize research initiative that grows a shared set of hybrids across dozens of field sites each season and releases the results, so that genotype, environment and management can be studied together rather than one at a time. This archive mirrors the two open prediction-competition packages drawn from that work. Each holds plot-level phenotypes for the training years, with grain yield in Mg per hectare at 15.5 percent moisture alongside stand count, plant and ear height, pollen and silk date, lodging counts, grain moisture and test weight; trial metadata with GPS coordinates, planting and harvest dates and management; soil chemistry per site; daily weather from NASA POWER for the full year and for the season window; genotypes for every hybrid as a VCF and as a numeric marker matrix; and a set of 654 pre-computed environmental covariates indexed by growth stage. The 2024 edition covers training years 2014 to 2023 against 22 held-out 2024 environments, and the 2022 edition covers 2014 to 2021 against the 2022 season. Both include the organizers' published observed values for their held-out year, so a model can be scored against a real season before it is submitted anywhere.\n\nThe crop is maize, Zea mays L., grown as multi-environment field trials of named hybrids, so the data suits genomic prediction, genotype by environment interaction, general and specific combining ability, reaction norms and other plant quantitative genetics work.","displayTitle":"Genomes to Fields: maize genotype by environment prediction competition data, 2014 to 2024","doi":"10.25739/78mn-4394","fuseMountBrowsable":true,"id":"55f04f5e-c111-5eee-901a-c1eb0abe5a8a","license":"PDDL-1.0","licenseEvidenceUrl":"https://api.datacite.org/dois/10.25739/78mn-4394","objectCountEstimate":2,"probeStatus":"ok","referenceUrl":"https://www.genomes2fields.org/resources/","region":"us-east-1","schemaCard":{"accessHints":[{"language":"python-anonymous","snippet":"import boto3\nfrom botocore import UNSIGNED\nfrom botocore.client import Config\ns3 = boto3.client('s3', region_name='us-east-1', config=Config(signature_version=UNSIGNED))\ns3.list_objects_v2(Bucket='scigantic-g2f-maize', Prefix='', Delimiter='/')"},{"language":"python-local","snippet":"open('/mnt/archive/prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv', 'rb')"}],"caveats":["Files are binary — use a domain-specific parser (e.g. metpy / pynexrad / pydicom) rather than text I/O."],"format":"binary","generatedAt":"2026-11-16T02:24:22.530Z","layout":{"example":"prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv","pattern":"{prefix}/{seg1}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"Env,Hybrid,Yield_Mg_ha\r\nDEH1_2022,B14A/OH43,\r\nDEH1_2022,B37/H95,\r\nDEH1_2022,B73/MO17,\r\nDEH1_2022,B73/PHN82,\r\nDEH1_2022,B73/TX779,\r\nDEH1_2022,B97/LH244,\r\nDEH1_2022,BGEM-0039-N/LH244,\r\nDEH1_2022,BGEM-0063-N/LH244,\r\nDEH1_2022,BGEM-0089-N/LH244,\r\nDEH1_2022,BGEM-0124-N/LH244,\r\nDEH1_2022,BGEM-0129-N/LH244,\r\nDEH1_2022,CG119/CG108,\r\nDEH1_2022,CG44/CGR01,\r\nDEH1_2022,DK3IIH6/LH244,\r\nDEH1_2022,DK78010/LH244,\r\nDEH1_2022,GEMN-0096/LH244,\r\nDEH1_2022,GEMN-0096_LH212HT_0001/LH244,\r\nDEH1_2022,GEMN-0096_LH212HT_0002/LH244,\r\n","encoding":"utf8","key":"prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv","truncatedAt":512},{"bytes":"Year,Env,Experiment_Code,Treatment,City,Farm,Field,Trial_ID (Assigned by collaborator for internal reference),\"Soil_Taxonomic_ID and horizon description, if known\",\"Weather_Station_Serial_Number (Last four digits, e.g. m2700s#####)\",Weather_Station_Latitude (in decimal numbers NOT DMS),Weather_Station_Longitude (in decimal numbers NOT DMS),Date_weather_station_placed,Date_weather_station_removed,Previous_Crop,Pre-plant_tillage_method(s),In-season_tillage_method(s),Type_of_planter (fluted cone; belt cone; ai","encoding":"utf8","key":"prediction-competition-2022/Testing_data/2_Testing_Meta_Data_2022.csv","truncatedAt":512}],"sampleObjects":["prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv","prediction-competition-2022/Testing_data/2_Testing_Meta_Data_2022.csv","prediction-competition-2022/Testing_data/3_Testing_Soil_Data_2022.csv","prediction-competition-2022/Testing_data/4_Testing_Weather_Data_2022.csv","prediction-competition-2022/Testing_data/6_Testing_EC_Data_2022.csv"],"sidecars":[{"bytes":"This release includes all the files used in the Genomes to Fields 2022 Maize Genotype by Environment Prediction Competition held from November 15, 2022, through January 15, 2023. The file named \"COMPETITION_DATA_README.docx\" is the exact file provided to the participants of the competition and contains all information related to the files in this release. The two exceptions are the files named \"Training_data/GenoDataSourcesWithUpdatedBioProject.txt\" and \"Testing_data/Test_Set_Observed_Values_ANSWER.csv\". \"Training_data/GenoDataSourcesWithUpdatedBioProject.txt\" is a copy of the \"GenoDataSources.txt\" file, including public source information unavailable during the competition. \"Testing_data/Test_Set_Observed_Values_ANSWER.csv\" contains the true observed values that were NOT provided to the participants but were used by the organizers to determine the accuracy of predictions submitted. The observed values file is in the same format as the template file named \"Testing_data/1_Submission_Template_2022.csv.\" Please note that some of the observations in the submission template file are not included in the observed values file; this is due to missing data that was not yet known to the organizers when the competition began, as not all trial location data had been reported at that time. For the competition, predictions of observations submitted using the template that did not have a corresponding observed value were dropped, and accuracy was only calculated using values with observations.\n\n\nFor general information, questions about the Genomes to Fields (G2F) initiative, The G x E Project and questions regarding released data set, please contact g2f@wisc.edu.\n\n\nOrganizing Committee:\nJacob Washburn: Research Geneticist, USDA-ARS.\nNatalia de Leon: Professor, University of Wisconsin-Madison.\nJames Holland: Research Geneticist, USDA-ARS.\nDavid Ertl: Iowa Corn Growers' Association.\nJoseph L. Gage: Assistant Professor, North Carolina State University.\nMaria Cinta Romay: Project Manager, Cornell University.\nDayane Cristina Lima: Researcher II, University of Wisconsin-Madison.\nJose Ignacio Varela: Postdoctoral Researcher, University of Wisconsin-Madison.\nQiuyue Chen: Postdoctoral Researcher, North Carolina State University.\n\n\nInventory:\n\n/readme.txt (this file)\n\n/COMPETITION_DATA_README.docx\n\t   Contains all information related to the files used in the competition.\n\n/Training_data: This directory contains 9 files\n\t1_Training_Trait_Data_2014_2021.csv\n\n\t2_Training_Meta_Data_2014_2021.csv\n\t   \t   Contains the basic metadata for the locations and years in the training set. \n\n\t3_Training_Soil_Data_2015_2021.csv\n\t   \t   Contains the basic soil data for the locations and years in the training set. \n\n\t4_Training_Weather_Data_2014_2021.csv\n\t   Contains the weather data downloaded from the NASA Power website (https://power.larc.nasa.gov/) for the locations and years in the training set. \n\n\t5_Genotype_Data_All_Years.vcf.zip\n\t   Contains all of the genotype data for both the training and testing sets. \n\n\t6_Training_EC_Data_2014_2021.csv\n\t   Contains 765 environmental covariates for the locations and years in the training set. \n\n\tAll_hybrid_names_info.csv\n\t   Information for hybrids presented in the dataset.\n\n\tGenoDataSources.txt\n\t   Contains genotype names, sample names, technology type, and public source when available.\n\n\tGenoDataSourcesWithUpdatedBioProject.txt\n\t   Copy of GenoDataSources.txt file including public source information not available during the competition.\n\n/Testing_data: This directory contains 6 files\n\t1_Submission_Template_2022.csv\n\t   This file contains the environments and hybrids for which participants submitted yield predictions.\n\n\t2_Testing_Meta_Data_2022.csv\n\t   Contains the basic metadata for the locations and years in the testing set. \n\n\t3_Testing_Soil_Data_2022.csv\n\t   Contains the basic soil data for the locations and years in the testing set. \n\n\t4_Testing_Weather_Data_2022.csv\n\t   Contains the weather data downloaded from the NASA Power website (https://power.larc.nasa.gov/) for the locations and years in the testing set. ","encoding":"utf8","key":"prediction-competition-2022/readme.txt","name":"readme.txt","size":4368,"truncated":true}],"sourceObjectKey":"prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv","starterCell":"with open('/mnt/archive/prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv', 'rb') as fh:\n    head = fh.read(1024)\nprint(len(head), head[:64])","starterSamplePath":"prediction-competition-2022/Testing_data/1_Submission_Template_2022.csv","totalObjects":"2 top-level prefixes"},"summary":"Genomes to Fields (G2F) is a public maize research initiative that grows a shared set of hybrids across dozens of field sites each season and releases the results, so that…","title":"Genomes to Fields: maize genotype by environment prediction competition data, 2014 to 2024","updatedAt":"2026-11-16T15:51:47.000Z","visibility":"public","websiteUrl":"https://www.maizegxeprediction.org/about/"}}