How to handle the different trait data of the same line?

I am trying to extract the Terra-REF trait data to run GWAS on biomass. So I extracted “aboveground_dry_biomass” but found that the values are quite different for the same line (see below).

• write.table(biomass[,c('cultivar', 'mean', 'raw_date', 'units')], file="biomass.txt", sep="\t")
• grep "PI156217" biomass.txt
    • "68"  "PI156217"  17810   "2017-09-12 00:00:00 -0500" "kg / ha"
    • "732" "PI156217"  30800   "2014-10-11 19:00:00 -0500" "kg / ha"
    • "1877"    "PI156217"  15200   "2018-08-01 02:00:00 -0500" "kg / ha"
    • "2094"    "PI156217"  23210   "2017-09-15 00:00:00 -0500" "kg / ha"
    • "3854"    "PI156217"  8540    "2018-07-31 02:00:00 -0500" "kg / ha”

Even for 2018, the values are 15200 vs 8540. I noticed that there are other biomass trait data. Do you think it is necessary to normalize these values to other traits or I can simply take the average for 2018? For GWAS, I need to compare traits against all BAP lines.

Hi Liya - thanks for your interest. Note that these results come from:

three different years (2014, 2017, 2018),
two different locations (PeeDee research center in SC and Maricopa Ag Center in AZ )
- (also note that the SC data are from Brenton et al 2016, and aren't part of the TERRA REF data)
two different treatments in 2017, one with drought treatment from Aug 1-14 and the other with drought treatment from Aug 15-30

Even then, it is surprising that the two values for 2018 are so different. @NewcombMaria can you confirm that these two measurements are correct?

For reference, I can access these results using the API: https://terraref.org/bety/api/v1/search?cultivar=PI156217&trait=aboveground_dry_biomass&key=YOURSECRETKEYHERE or database: select sitename, city, cultivar, treatment, mean, raw_date, units from traits_and_yields_view where cultivar = 'PI156217' and trait = 'aboveground_dry_biomass';

sitename	city	cultivar	treatment	mean	raw_date	units
MAC Field Scanner Season 4 Range 4 Column 7	Maricopa	PI156217	MAC Season 4: BAP water-deficit stress Aug 1-14	17810	2017-09-12 05:00:00	kg / ha
Clemson University Pee Dee Research and Education Center	Florence	PI156217	2014 Clemson BAP Phenotyping Trials	30800	2014-10-12 00:00:00	kg / ha
MAC Field Scanner Season 6 Range 34 Column 8	Maricopa	PI156217	MAC Season 6: Sorghum	15200	2018-08-01 07:00:00	kg / ha
MAC Field Scanner Season 4 Range 34 Column 10	Maricopa	PI156217	MAC Season 4: BAP water-deficit stress Aug 15-30	23210	2017-09-15 05:00:00	kg / ha
MAC Field Scanner Season 6 Range 6 Column 10	Maricopa	PI156217	MAC Season 6: Sorghum	8540	2018-07-31 07:00:00	kg / ha

terraref / reference-data

How to handle the different trait data of the same line? #290