Comments from curator
Phenotyping data comes from the
S1 File (
pone.0160857.s001.xlsx) in the
Supporting Information of the paper:
-
10 Line Raw Data: wet weight of each weighed group of pupae (2,837 groups) of the "10 line study" (DGRP_153, DGRP_440, DGRP_748, DGRP_787, DGRP_801, DGRP_802, DGRP_805, DGRP_900, DGRP_907 and DGRP_911), used for raw dataset 1.
-
4 Line Raw Data: the same for the "4 line study" (DGRP_153, DGRP_440, DGRP_748 and DGRP_911), with a "Fall" and a "Spring" replicate. The 1,239 Fall rows are the 10 line study rows of these four lines (same values), so only the 1,718 Spring rows are added, as raw dataset 2 (phenotypes with the suffix _Spr).
-
10 Line Weight Mean SE and
4 Line Weight Mean SE: the authors' number of groups, mean and standard error of the weight per pupa for each line, sex, generation and treatment. The means computed by DGRPool from the raw datasets reproduce them exactly (160/160 and 64/64), so no summary dataset is uploaded.
Design: flies were kept at 25 °C, 12:12 h light:dark and 50% humidity. In the parental generation (P), vials of 50 first-instar larvae of each line were reared on a normal cornmeal-molasses diet (Control) or on the same diet with 3% coconut oil (high fat, HF). P adults of the same line were then crossed to found three lineages: Control (Control females × Control males), maternal ancestor
MA (HF females × Control males) and paternal ancestor
PA (Control females × HF males). F1 adults were mated within their lineage to produce the F2, and all F1 and F2 larvae were reared on normal food. Pupae were collected within 12 h of eclosion (when they can be sexed), stored in Ringer's solution at -20 °C, then cleaned, sexed, patted dry and weighed in groups of three (wet weight, Mettler Toledo XS105 balance). All crosses are made within a DGRP line, so the phenotypes are those of the DGRP line itself.
Raw datasets: one value is one weighed group of pupae of one line, sex, generation and treatment. The value is the weight of the group divided by its number of pupae, i.e. the mean wet weight of one pupa (mg), as in the authors' mean sheets. Most groups have three pupae (2,736 of the 2,837 groups of the 10 line study; the others have 1, 2 or 4); the number is given in the obsolete
NPupae_* columns. Groups of different conditions are independent, so they are stacked: row k of a line and sex holds the k-th group of each condition. The phenotypes are named
PupaWt_<generation>_<treatment>:
-
PupaWt_P_Ctrl and
PupaWt_P_HF: P generation reared on normal and on high-fat food;
-
PupaWt_F1_Ctrl,
PupaWt_F1_MA,
PupaWt_F1_PA,
PupaWt_F2_Ctrl,
PupaWt_F2_MA and
PupaWt_F2_PA: F1 and F2 descendants of the Control, MA and PA lineages, all reared on normal food;
- suffix
_Spr: Spring replicate of the 4 line study (raw dataset 2).
The authors analysed the transgenerational effects with ANOVAs (genotype, treatment, generation and sex) and correlated the treatment effects of ancestors and descendants (Table 3, using differences between the treated and Control means of a line, sex and generation). No per-line differences were published and there was no GWAS. The effect of an ancestral treatment can be derived from the uploaded phenotypes, e.g. the mean of PupaWt_F2_MA minus the mean of PupaWt_F2_Ctrl for the same line and sex.
Of note, two corrections to the Spring rows were needed to reproduce the authors' means:
- DGRP_748, F1 Control females: one group of 3.62 mg is recorded as a single pupa (3.62 mg per pupa, about 2.7 times the other groups of this line and sex). The authors' mean (28 groups, 1.3276 mg) is reproduced only with three pupae, so it is uploaded as a group of three (1.2067 mg per pupa).
- DGRP_440, F2 PA: the sheet has two separate blocks of groups for each sex (33 and 25 female groups, 27 and 25 male groups). The authors' means (25 groups per sex) are those of the second block, which is uploaded as PupaWt_F2_PA_Spr. The first block, not used in the authors' means and of unknown origin, is kept as the obsolete phenotype
PupaWt_F2_PA_SprB (mean 1.415 mg in females and 1.098 mg in males, vs 1.476 and 1.160 mg in the second block).
Not uploaded: the "3 line study" (sheets "3 Line ..."), i.e. larval triglyceride, trehalose and protein levels, egg size and individual pupal weights of three lines of the
Drosophila Synthetic Population Resource (DSPR 11291, 12062 and 22271), which are not DGRP lines; and the variance partitions of the ANOVAs (sheets "... variance explained").