Study

Curated
Leiva et al., 2023
Intraspecific variation of heat tolerance in a model ectotherm: The role of oxygen, cell size and body size
[10.1111/1365-2435.14485]

Description

Heat tolerance (survival time at a constant 36-39 °C) of single virgin females and males of 20 DGRP lines with small or large wing cells, under normoxia and, for 6 lines, hypoxia, plus fresh mass, wing and cell size and interocular distance. Hypoxia shortened survival; larger flies survived longer.

Comments from curator

Phenotyping data comes from the authors' data and code repository cited in the Data Availability Statement of the paper: GitHub felixpleiva/TDT_DGRP_lines, archived on Zenodo (10.5281/zenodo.5120028; version 3.0 of 2023-11-22, 10.5281/zenodo.5244109, was used). Its Data folder holds three Excel files exported from the authors' database:
- results on TDT available.xlsx: one row per reading of the survival time of one fly in the heat assays, used for raw dataset 1.
- results on CS available.xlsx: body-size traits of single flies (three repeated measurements of the wing and of the interocular distance per fly), used for raw dataset 2.
- fresh mass and dry mass correlation 4 lines 20191206.xlsx: fresh and dry mass of single flies of 4 lines, used for raw dataset 3.
The authors' R scripts (R folder) were read to reproduce their processing. The 20 lines are given as Bloomington stock numbers, mapped to DGRP lines (they agree with the DGRP numbers of the GENOTYPE column).

Raw dataset 1: heat tolerance (thermal death time curves). Virgin flies of both sexes were kept as adults at 25 °C, then tested individually in sealed glass vials immersed in a water bath at a constant 36, 37, 38 or 39 °C. Time-lapse pictures (one every 10 s) were turned into reversed videos, and the survival time is the time at which the fly ceased to move. Normoxia (21 kPa O2, Series 1): 1,686 flies of the 20 lines, aged 4 to 8 days (5 to 8 in the paper). Hypoxia (10 kPa O2, Series 2): 6 lines (3 small-celled: DGRP_313, DGRP_324, DGRP_843; 3 large-celled: DGRP_426, DGRP_712, DGRP_732) kept for 5 to 8 days after emergence in boxes flushed with 10% O2 and tested in septum-capped vials flushed with 10% O2. As in the authors' scripts, the normoxic flies of these 6 lines (538 flies, tested in 2018 and 2019) are pooled with Series 1. Each phenotype is one temperature and oxygen condition (SurvTime_<temperature>C_<oxygen>kPa), in minutes, one value per fly; the flies of the different conditions are different flies, stacked in a compact layout (row k of a line and sex holds the k-th fly of each condition). 226 flies were scored by two readers: their value is the geometric mean of the two readings, which is what the authors' median of the log10 times gives. These values reproduce the per-fly log10 survival times of the authors' files (Outputs 4.1.1 and 4.1.2) for all 2,088 flies they contain (differences below 1e-14). The authors analysed the median of the ~10 flies of each line, sex and condition (on the log10 scale); the median computed by DGRPool is the same when the number of flies is odd and differs by at most 1.4% otherwise.
Changes and remarks:
- The 20 hypoxic flies of DGRP_324 tested at 36 °C (run of 2019-07-14) have no recorded video duration, so the authors' R script (aggregate() drops rows with a missing value) left them out: the paper has 402 hypoxic flies and no DGRP_324 data at 36 °C under hypoxia. They are included here (422 hypoxic flies), so SurvTime_36C_10kPa of DGRP_324 is not in the paper's analyses.
- For one fly (DGRP_361 male, 36 °C, normoxia), the second reading (189:57) has no video duration and is longer than the 121-min video; as in the authors' script, only the other reading (90:32) is used.
- One fly (DGRP_088 female, 36 °C, normoxia) has a survival time (201.4 min) longer than its recorded video duration (154.35 min); it is kept as published.
- DGRP_732 was not tested at 37 °C under hypoxia, DGRP_361 females were not tested at 37 and 38 °C under normoxia, and only one DGRP_440 female was tested at 36 °C under normoxia.
The thermal death time curve parameters (CTmax and thermal sensitivity z, from the regression of log10 survival time on test temperature per line and sex) are not given per line in the paper and were not computed here. The same data were reanalysed in a companion preprint (Leiva et al. 2024, EcoEvoRxiv 10.32942/X20C9T, "Genetic variation of heat tolerance in a model ectotherm: an approach using thermal death time curves"), which estimates the heritability of CTmax and z. The mean ± SD survival times it reports at 36 and 39 °C (e.g. 150 ± 47 min for females at 36 °C) are reproduced by the flies of this dataset, except the SD of males at 39 °C (5.6 in the preprint, 6.77 over flies, 5.57 over line means).

Raw dataset 2: body size and cell size (Figure S1). One row per fly: 389 flies aged 5 days, 8 to 11 per line and sex, a different set of flies from the heat assays. FreshMass (mg) was weighed on CO2-anaesthetised flies, which were then stored in glycerol:ethanol. WingArea (µm²) is the area enclosed by seven landmarks of one wing, mean of the three repeated measurements. CellArea (µm²) is the counting area divided by the number of trichomes in it (one trichome per cell), and CellNumber is WingArea / CellArea (the authors' formulas, script 1.0; they reproduce the authors' per-fly values). IOD (interocular distance, µm) is the mean of the three repeated measurements. IOD was not measured in 6 lines (DGRP_057, DGRP_195, DGRP_338, DGRP_361, DGRP_399, DGRP_892) and in 4 single flies, and it is set to NA for DGRP_732, which the authors excluded because of a problem with the scale of the stereomicroscope. The authors predicted the IOD of these 7 lines from fresh mass (script 2.0); these predicted values are not uploaded. The paper reports 384 flies: the authors' script dropped the 4 flies without IOD and one DGRP_426 female; they are kept here, so the line medians computed by DGRPool differ slightly (at most 1.5%) from those the authors used as covariates for 5 of the 40 line/sex combinations.

Raw dataset 3: fresh and dry mass. One row per fly (118 flies of DGRP_313, DGRP_380, DGRP_426 and DGRP_732, 5 days old, kept at 25 °C and 21 kPa O2, measured on 2019-12-06): FreshMass_DMset and DryMass, in mg. These data are not used in the paper; the authors' script 9.0 only plots the correlation between dry and fresh mass (leaving out DGRP_732 without giving a reason). Two empty wells were dropped; one fly has no fresh mass.

The paper also cites the related study of the same group on the plasticity of body and cell size (Leiva, Boerrigter & Verberk 2023, Functional Ecology, 10.1111/1365-2435.14294), which is DGRPool study 141.

Datasets

Dataset Phenotypes DGRP lines
Raw dataset #1 8 phenotypes
SurvTime_36C_10kPa SurvTime_36C_21kPa SurvTime_37C_10kPa SurvTime_37C_21kPa SurvTime_38C_10kPa SurvTime_38C_21kPa SurvTime_39C_10kPa SurvTime_39C_21kPa
20 DGRP lines
DGRP_057
DGRP_088
DGRP_129
DGRP_195
DGRP_313
DGRP_324
DGRP_338
DGRP_361
DGRP_380
DGRP_385
DGRP_399
DGRP_426
DGRP_440
DGRP_441
DGRP_555
DGRP_712
DGRP_732
DGRP_843
DGRP_849
DGRP_892
Download
Raw dataset #2 5 phenotypes
CellArea CellNumber FreshMass IOD WingArea
20 DGRP lines
DGRP_057
DGRP_088
DGRP_129
DGRP_195
DGRP_313
DGRP_324
DGRP_338
DGRP_361
DGRP_380
DGRP_385
DGRP_399
DGRP_426
DGRP_440
DGRP_441
DGRP_555
DGRP_712
DGRP_732
DGRP_843
DGRP_849
DGRP_892
Download
Raw dataset #3 2 phenotypes
DryMass FreshMass_DMset
4 DGRP lines
DGRP_313
DGRP_380
DGRP_426
DGRP_732
Download
Back to studies