Comments from curator
Phenotyping data come from the authors'
Dryad deposit (
doi:10.5061/dryad.2hg1785, CC0):
Choice.csv,
NoChoice.csv,
P1.csv and
P2.csv, one row per trial (focal male).
Aggression.csv (preliminary aggression assay on wild-caught, non-DGRP males) is not used. The paper itself has no per-line values (variance components only, Tables 1 and 2).
Design. The authors used 28 randomly chosen DGRP lines and crossed each one to a single "standardized reference line" to avoid inbreeding effects: the resulting "clonal hybrids" share the maternal haplotype of the reference line and carry the haploid genome of their DGRP father.
All phenotypes are F1 phenotypes, stored under the DGRP father line (the files code the hybrids as M<RAL number>, e.g. M26 = DGRP_026). The reference line is not named in the paper, nor in the two later papers of the same authors on 28 DGRP hybrids made the same way (
Filice, Bhargava & Dukas 2020, "a randomly selected standardized reference line";
Filice, Bhargava & Dukas 2021), hence the name suffix
_xRef. It is probably DGRP_83, the "randomly-chosen standard line" of the lab's earlier DGRP hybrids (
Scott, Dworkin & Dukas 2018, cited for this hybrid design by the 2021 paper) and of Baxter et al. 2023 (DGRPool study 213), and DGRP_83 is not among the 28 fathers, but the authors do not say so. The F1 males carry the X chromosome of the reference line. Focal F1 males (4 days old) were randomly assigned to two treatments: 4 h in an aggression arena with a young (1-day-old, docile) male of the hyperaggressive "bully" population (
winners, _W) or with a mature (5-day-old) bully male (
losers, _L), which was about 18 times more aggressive. They were then tested in four assays: (1)
choice: with a competitor male and a female of the outbred IV population, scored for which male mated (14 days); (2)
no-choice: with an IV female alone, mating latency and duration (12 days); (3)
sperm defence (P1): the focal male mated a virgin IV-bw (brown-eyed) female, whose offspring of the first night measure fecundity after a single mating; she was offered an IV-bw male the next day, and the red- (focal) and brown-eyed offspring of the 24 h after remating give the focal male's paternity share; (4)
sperm offence (P2): the same with the focal male mating second. Assays 3 and 4 used a random subset of 6 of the 28 hybrids (3 replicate days of 10 males per hybrid and treatment). Winners mated more often in the choice test and 1.5 times faster in the no-choice test, while losers mated longer, induced more offspring after a single mating and had a higher P1 paternity share; P2 did not differ. The hybrid effect was significant for mating duration and P2 only (GLMMs).
Datasets (sex M: the focal F1 male; winner and loser males are different flies, stacked compactly: row k of a hybrid and day holds its k-th winner and k-th loser male in file order; times in seconds, values as in the files):
-
raw_1 (choice test): 390 rows, 28 lines;
ChMatSucc_W_xRef /
_L_xRef (column
win: 1 = the focal male mated, 0 = the competitor mated; the line mean is the proportion of trials won) and the obsolete test day
ChDay_xRef. The column
loss (= 1 - win) is not uploaded.
-
raw_2 (no-choice test): 283 rows, 28 lines;
NCLat and
NCDur (mating latency and duration) _W_xRef / _L_xRef and the obsolete test day
NCDay_xRef (coded W1-W12 in the file).
-
raw_3 (P1): 106 rows, 6 lines; latency and duration of the focal male's mating (
P1Lat1,
P1Dur1), single-mating fecundity (
P1Fec), latency and duration of the female's remating with the IV-bw male (
P1Lat2,
P1Dur2), offspring sired by the focal and by the IV-bw male after remating (
P1OffFoc,
P1OffBw) and the authors' paternity share of the focal male, red / (red + brown) offspring (
P1Share), each _W_xRef / _L_xRef, and the obsolete replicate day
P1Day_xRef.
-
raw_4 (P2): 97 rows, 6 lines; the same columns with the matings in the female's order: IV-bw male first (
P2Lat1,
P2Dur1), focal male second (
P2Lat2,
P2Dur2), then
P2OffFoc,
P2OffBw,
P2Share and
P2Day_xRef.
Lines: DGRP_026, DGRP_028, DGRP_091, DGRP_129, DGRP_195, DGRP_217, DGRP_239, DGRP_315, DGRP_357, DGRP_367, DGRP_371, DGRP_373, DGRP_379, DGRP_392, DGRP_395, DGRP_399, DGRP_439, DGRP_443, DGRP_491, DGRP_492, DGRP_502, DGRP_508, DGRP_596, DGRP_703, DGRP_765, DGRP_843, DGRP_900, DGRP_907; assays 3-4: DGRP_091, DGRP_129, DGRP_239, DGRP_379, DGRP_399, DGRP_907.
Data handling. The files hold only trials with a mating (choice: 333 winner and 336 loser trials; no-choice: 258 and 276, as in the paper) and, for P1 and P2, only the females that remated (P1: 85 and 89; P2: 90 and 65): trials without a mating and trials that could not be run are absent. Nothing was computed; the winner-loser difference of a line can be derived from its _W and _L means.
Of note:- In the choice test, the files give one male per hybrid, treatment and day in most cases, but two in 28 hybrid x day groups (hence extra rows). The proportions of trials won are 0.589 (winners) and 0.470 (losers); the paper gives 0.58 and 0.48.
- P1: the paper reports 89 winner and 85 loser females that remated, the file has 85 winner ("young" bully) and 89 loser ("mature") trials. The treatment coding of the file reproduces the paper's results (winners: 8.4% lower P1 paternity share, 0.111 vs 0.195, and fewer offspring after a single mating, 46.2 vs 51.6), so the counts of the text are probably swapped.
- P2: one negative mating duration (-77 s, IV-bw male's mating with the female of a DGRP_129 loser trial, P2Dur1_L_xRef) is kept as in the file.
- The data reproduce the paper's other numbers: no-choice latency of losers 1.51 times that of winners, losers mating longer (883 vs 814 s); mean paternity share of the IV-bw male in P1 84.7% (paper: 84.7%) and of the focal male in P2 92.4% (paper: 92.4%).
Paper:
Proc. R. Soc. B 286:20182838 (also on
PMC).