Comments from curator
Phenotyping data comes from the U.S. EPA Science Hub dataset of the paper (
doi:10.23719/1419253, file
HerrDavid_A-qnkv_DATA_20171221.xlsx), sheet
GWAS input data: the per-line values of the 123 DGRP lines used for the GWAS (lines given as bare numbers, e.g. 21 = DGRP_021). Only these per-line values were published (no per-fly data), so the study has a summary dataset only. All values are from males (sex M); females were not tested. The
journal supplement (also on
PMC) holds the GWAS results and gene lists, which are also in the EPA workbook.
Design: males of each line, 42-48 h old, were loaded individually under CO
2 into the glass tubes of
Drosophila Activity Monitors (DAM2, TriKinetics), sealed at both ends by gas-distribution manifolds, at 25 °C in dim light (6-8 lux). Photobeam breaks were summed per 10-min bin for 20 h: 1 h of recovery from anaesthesia (pre), 4 h of exposure (dose) to 750 ppm toluene (close to the EC20 of males of the Flyland population) or to clean air, then 15 h after the exposure (post), late in which the flies died. Each day, 8 flies of a line received toluene and 8 paired flies received air; each set of 16 lines was tested at least 3 times, giving 11-60 flies per line in air and 17-61 in toluene (the per-line numbers are not published).
Phenotypes (values copied as in the file, mostly 3 significant digits; suffix _Air = clean air control, _Tol = toluene):
-
mn_ActPre,
mn_ActDose,
mn_ActPost and
mn_ActTotal: activity during the acclimation hour, the 4-h exposure, the 15 h after exposure and the whole 20 h (columns AUC.pre, AUC.dose, AUC.post and AUC.total: area under the curve of the counts per 10-min bin, i.e. the total photobeam breaks per fly in the period; AUC.total = pre + dose + post for all lines). The paper's GWAS (DGRP2 web portal) used the activity during exposure in air and in toluene, and their difference.
-
mn_ActOnset and
mn_ActOffset: change in activity at the onset and at the offset of the exposure (columns delta.on and delta.off). They are not described or used in the paper; the companion paper by Tatum-Gibbs et al. (2015,
doi:10.1016/j.ntt.2014.11.008, on the Flyland population) reports transient increases in activity at the onset and offset of toluene exposure.
-
mn_DeathTime: time to death, in minutes from the start of the recording, at 10-min resolution. 1200 means that the flies were still alive at the end of the 20-h recording (14 lines in air, 20 in toluene); this value is kept, as the authors used it in their differences. How the per-line value was derived from the flies is not described.
-
_Diff: toluene minus air difference of each measure, as published per line in the file (columns Tol-Air; they equal the difference of the two uploaded columns).
mn_ActDose_Diff is the narcotic effect of toluene and the main GWAS phenotype of the paper (the legend of Supplementary Table S1 defines sensitivity as air minus toluene, the text and the file as toluene minus air).
-
sd_Act10m (_Pre, _Dose, _Post, _Total): columns SD.AUC.* of the file, not described in the paper. They are not standard deviations of the period totals: SD.AUC.total equals the average of the pre, dose and post values weighted by their number of 10-min bins (6, 24 and 90) for all 123 lines, so they are standard deviations of the counts per 10-min bin, averaged over the bins of the period (most likely across the flies of the line).
Not uploaded: the four "Mean SD" columns (Excel formulas averaging the air and toluene SD columns), and the sheet "Summary Stats", labelled preliminary in the file: 112 lines, 16 re-runs coded 1000xxx and the Flyland hybrid population (not a DGRP line); its values differ from the final ones for about 40% of the lines, with fewer flies. DGRP_364 and DGRP_489 are only in this preliminary sheet (the file's table of contents mentions a list of lines that died and were left out of the GWAS, in a sheet missing from the file). The full GWAS results are on the DGRP2 web portal.
Checks against the paper: toluene suppressed activity during exposure in 104 of the 123 lines and increased activity after exposure in 75 lines, as in the paper. Toluene increased the time to death in 72 lines (9 equal, 42 shorter), whereas the paper says 77.