G2F RESOURCES

This repository provides resources (data and pipelines) linked to data generated by the GxE initiative of the Genomes to Fields project.

Daily weather data

Source	Output-file	Inputs	Script	Data dictionary
NASA	NASA_WeatherData.csv	info_loc.csv	link	link
G2F	g2f_WeatherData.csv	G2F_weather_2018.csv G2F_weather_2019.csv NASA_WeatherData.csv	link	link

The Consensus weather data combines data from G2F with data from NASA. NASA data was used for variables not recorded by G2F and to fill records from G2F that were either missing or appeared to be wrong. For further information about each of these networks please see the corresponding "mdfiles".

• Weather data maintained by Fernando Aguate ([email protected]), Marco Lopez-Cruz ([email protected]), and Gustavo de los Campos ([email protected])

Soil data

Source	Output-file	Inputs	Script	Data dictionary
G2F	g2f_SoilData.csv	g2f_2018_soil_data.csv g2f_2019_soil_data.csv	link	link
USDA	USDA_SoilData.csv	info_loc.csv	link	link
Consensus G2F and USDA	SoilData.csv	g2f_SoilData.csv USDA_SoilData.csv	link	link

• Soil data maintained by Jim Holland ([email protected])

Phenotypes

What?	Input	Script	Output	Description
Fortmatted phenotypes	repository G2F site	link	phenotype.csv	link
Processed phenotypes	phenotype.csv info_loc.csv loc_GDD_clusters.csv	link	clean_pheno_yearloc_means.csv clean_pheno_raw.csv

Analysis pipelines

Predicting Growing Degree Days (GDD) needed to silking for each hybrid

What?	Output	Inputs	Script
GDD to silking	data_model.rdata	ConsensusData.csv phenotype.csv	link
GDD to silking/harvest	GDD_hat_by_hybrids.csv	phenotype.csv	link

Run APSIM Crop model to generate environmental covariates

Weather data for APSIM can be downloaded with this script

Previous results can be found here

What?	Input	Script	Output	Description
Simulated EC	clean_pheno_yearloc_means.csv	link	simulated_EC_V1.csv	link
EC by stage	simulated_EC_V1.csv	link	clean_EC_by_stage.csv output

Variance components analysis

What?	Input	Script	Output	Description
ANOVA	clean_pheno_raw.csv clean_EC_by_stage.csv	link	files ANOVA
Sequential ANOVA	clean_pheno_raw.csv clean_EC_by_stage.csv	link	files sequential ANOVA
Summary ANOVA	files ANOVA files sequential ANOVA	link	output

Validation

ECs were assesssed for its ability to predict year-location means using leave-one-out (LOO) cross-validation assigning each year-location (YL) to folds. The ability to predict years performance are also evaluated.

What?	Input	Script	Output	Description
LOO YL replicates	clean_pheno_raw.csv clean_EC_by_stage.csv	link	files LOO YL
LOO YL means LOO YEAR means	clean_pheno_yearloc_means.csv clean_EC_by_stage.csv	link	files LOO YL means files LOO YEAR means
Summary LOO	files LOO YL files LOO YL means files LOO YEAR means	link	output