Study

Curated
Leiva et al., 2023
The role of cell size in shaping responses to oxygen and temperature in fruit flies
[10.1111/1365-2435.14294]

Description

Fresh mass, wing area, wing cell area and cell number of individual flies of 6 DGRP lines differing in cell size, reared at 17 or 25 °C under hypoxia (10 kPa O2) or normoxia (21 kPa). Size decreased with warming more strongly under hypoxia and in larger-celled lines.

Comments from curator

Phenotyping data comes from the authors' repository (GitHub, archived on Zenodo), file Data/body size related traits 20191122.xlsx (one row per fly, 477 flies). It was checked against the authors' analysed file Outputs/1.1.1. Data to fit the models.csv and their script R/1.0. Data wrangling.R.

The six DGRP lines were chosen for their wing cell size in Leiva et al. 2024 (Functional Ecology 38:439, DGRPool study 140): small-celled DGRP_313 and DGRP_324, medium-celled DGRP_129 and DGRP_843, large-celled DGRP_426 and DGRP_732 (Bloomington stocks 25180, 25182, 28141, 28247, 28196 and 25203 in the source). Flies were reared for one generation from egg to adult at 17 °C or 25 °C, under hypoxia (10 kPa O2) or normoxia (21 kPa O2), and kept as virgin adults for 5-8 days in the same condition before measurement. The condition is the suffix of each phenotype name (e.g. FreshMass_17C_10kPa).

The raw dataset has one row per fly, with four traits: fresh mass (mg), wing area (µm², area within seven landmarks), wing cell area (µm², area of a ~7850 µm² circle on the wing divided by the number of trichomes counted in it, one trichome per cell) and the estimated number of cells in the wing (wing area / cell area). Cell area and cell number were computed with the authors' formulas, at full precision. The four conditions are independent groups of flies, stacked in a compact layout: row k holds the k-th fly of each condition for a given line and sex, so values on the same row but in different conditions do not come from the same fly. The authors' missing-value code -9999 was set to NA. The storage plate and well, the wing side (left for 425 flies, right for 40) and the authors' log10-transformed columns were not uploaded.

The authors analysed only the flies with a trichome count: the 31 flies without one were excluded from all their analyses, fresh mass included. Their fresh mass (31 flies) and wing area (19 flies) were measured and are included here, but they are not in the paper; these are exactly the flies with an NA cell area. As a result, the DGRPool means of FreshMass and WingArea differ slightly from the means of the analysed flies for some line, sex and condition groups (by up to 0.018 mg and 8,030 µm²). CellArea and CellNumber are identical to the paper's data.

No flies of DGRP_129 were measured at 17 °C, and no females of DGRP_843 at 25 °C / 21 kPa. For this missing control group, the authors added one row ("new_fly") to their analysed file, whose values are the medians of 10 DGRP_843 females measured in Leiva et al. 2024 (DGRPool study 140). It is not a fly of this experiment and was not uploaded. The paper's 447 flies (221 females, 226 males) include this row, so 446 measured flies were analysed.

Figures 3-6 of the paper express fresh mass, wing area and cell area as a % change relative to the control condition: (value / median of the same line and sex at 25 °C and 21 kPa - 1) × 100. These per-fly derived values were not uploaded. They can be recomputed from the uploaded phenotypes using the medians of the flies analysed in the paper (those with a cell area), with the study 140 median for DGRP_843 females.

6 DGRP lines 6 genotyped

DGRP_129 DGRP_313 DGRP_324 DGRP_426 DGRP_732 DGRP_843
Back to studies