Comments from curator
Phenotyping data come from the authors' Figshare deposit (
10.17045/sthlmuni.13211204, version 1), cited in the paper's Data Availability Statement:
FLUCT_mating_fertility.txt (mating success and fertility) and
FLUCT_productivity_sexratio.txt (offspring counts), one row per focal DGRP male, i.e. per mating pair kept in a vial. The two files were merged by line, thermal condition and vial number into raw dataset 1.
Design: groups of about 50 eggs of each DGRP line developed, from egg to adult emergence, under one of four thermal conditions:
Constant 25 °C,
Fluctuating 25 °C (daily cycle between 21 and 29 °C),
Constant 29 °C and
Fluctuating 29 °C (daily cycle between 25 and 33 °C). In the fluctuating regimes, the high and low temperatures were each held for 6 h and separated by 6-h ramps. Virgin males were collected within 3 h of eclosion and kept singly at 25 °C. The next day, each focal male was paired with one virgin Canton-S female (1-5 days old, reared at 25 °C) for 3 days at 25 °C. Mating was observed during the first 8 h after pairing, the presence of larvae was scored 2 days after the pair was removed, and the adult offspring were counted 12 days after pairing and at the end of eclosion (14-18 days). The experiment was run in a single block on 30 lines, with about 15-20 males per line and condition. The lines were chosen as "high" lines (male fertility little affected by a high developmental temperature) or "low" lines (male fertility declining steeply at 29 °C) from the screen of Zwoinska et al. 2020 (
Front. Genet. 11:573,
DGRPool study 194):
- High lines: DGRP_059, 208, 227, 228, 256, 360, 362, 374, 375, 379, 439, 440, 584, 822, 861.
- Low lines: DGRP_021, 195, 229, 235, 319, 352, 365, 409, 437, 491, 530, 627, 808, 819, 900.
This line status is a property of the lines, not a phenotype of this study, so it was not uploaded.
Raw dataset 1 (1,894 focal males): one value per focal male.
The sex column is the sex of the focal DGRP fly (M); the offspring counts are those of the pair. For each developmental condition (suffix _Const25, _Fluct25, _Const29 or _Fluct29):
- Mating: 1 if mating was observed within the first 8 h after pairing, 0 if not (column Mating).
- Fertility: 1 if larvae were present, 0 if not (column Larvae).
- Productivity: total number of adult offspring of the pair (sum of the two counts; column Productivity).
- Daughters and Sons: number of adult daughters and sons (columns Daughters and Sons).
The four conditions are independent groups of males: values on the same row but in different condition columns come from different males. Row k of a line holds the k-th vial (by vial number) of each condition. The day of egg collection (EggCollDay, 1-3) and the day of pairing (PairingDay, 1-5) of each male are kept as obsolete design phenotypes.
Not uploaded:
- The DGRP line annotations repeated in the files (line status,
Wolbachia status, mitochondrial haplotype from Bevers et al. 2019, inversion karyotypes).
- The offspring sex ratio, which the paper analyses as the proportion of sons (binomial model of sons vs daughters). It is derived from the uploaded counts: the pooled sex ratio of a line is mn_Sons / (mn_Sons + mn_Daughters) from the DGRPool means of the same condition.
- The IGV column (absolute deviation of the productivity of each vial from the median of its line and condition, given by the authors as a measure of intra-genotypic variability). It is not used in the paper and is derived from Productivity. Moreover, in 6 of the 120 line x condition groups (85 vials: DGRP_021, 491, 530 and 900 at Fluctuating 25 °C, DGRP_235 at Fluctuating 29 °C and DGRP_584 at Constant 29 °C), the deposited values are not deviations from the median of the group.
The pooled means of the 15 high and 15 low lines computed from this dataset reproduce Table 1 of the paper: for mating success, fertility and productivity, 83 of the 88 published values (mean, SE, average number of males and of zeros per line) are identical at the published precision, and the other 5 differ by one unit in the last digit (rounding). The sex ratios of Table 1 (pooled proportion of sons) are reproduced from the Daughters and Sons counts for all 8 groups. The authors analysed mating success and fertility as binary traits and productivity after a Box-Cox transformation, in mixed models with thermal condition and line status as fixed effects and line as a random effect, and estimated the broad-sense heritability and evolvability of each trait per condition. No GWAS was done.
Missing values and inconsistencies (kept as published):
- Sample sizes vary: DGRP_021 has no male at Constant 25 °C and only 1-9 in the other conditions, and DGRP_627 (1), DGRP_365 (6), DGRP_409 (6) and DGRP_530 (7) have few males at Fluctuating 29 °C.
- 9 males have mating and fertility scores but no offspring counts (Productivity, Daughters and Sons NA): DGRP_195 and DGRP_352 (2) at Fluctuating 25 °C, DGRP_256 (5) and DGRP_439 at Constant 29 °C.
- 7 vials have a productivity but no daughters and sons counts (NA): DGRP_229 (5) and DGRP_319 at Constant 25 °C, DGRP_627 at Fluctuating 25 °C.
- In 2 vials, daughters + sons differ from productivity: DGRP_360 at Constant 25 °C (17 + 22 vs 40) and DGRP_375 at Constant 29 °C (13 + 9 vs 37).
- 34 vials scored without larvae produced 1 to 47 adult offspring, and 10 vials with larvae produced no adult offspring.
- Many males not observed mating within 8 h were fertile (603 males), because the pairs stayed together for 3 days.
Of note, the same lab measured the same traits on 40 lines (including these 30) developed at a constant 25, 27 or 29 °C in another experiment (
DGRPool study 58, Rodrigues et al. 2022, J. Anim. Ecol. 91:1119); the data of the two studies are different.