#PAGE_PARAMS# #ADS_HEAD_SCRIPTS# #MICRODATA#

Natural variation in the regulation of neurodevelopmental genes modifies flight performance in Drosophila


Authors: Adam N. Spierer aff001;  Jim A. Mossman aff001;  Samuel Pattillo Smith aff001;  Lorin Crawford aff002;  Sohini Ramachandran aff001;  David M. Rand aff001
Authors place of work: Department of Ecology and Evolutionary Biology, Brown University, Providence, Rhode Island, United States of America aff001;  Center for Computational Molecular Biology, Brown University, Providence, Rhode Island, United States of America aff002;  Microsoft Research New England, Cambridge, Massachusetts, United States of America aff003
Published in the journal: Natural variation in the regulation of neurodevelopmental genes modifies flight performance in Drosophila. PLoS Genet 17(3): e1008887. doi:10.1371/journal.pgen.1008887
Category: Research Article
doi: https://doi.org/10.1371/journal.pgen.1008887

Summary

The winged insects of the order Diptera are colloquially named for their most recognizable phenotype: flight. These insects rely on flight for a number of important life history traits, such as dispersal, foraging, and courtship. Despite the importance of flight, relatively little is known about the genetic architecture of flight performance. Accordingly, we sought to uncover the genetic modifiers of flight using a measure of flies’ reaction and response to an abrupt drop in a vertical flight column. We conducted a genome wide association study (GWAS) using 197 of the Drosophila Genetic Reference Panel (DGRP) lines, and identified a combination of additive and marginal variants, epistatic interactions, whole genes, and enrichment across interaction networks. Egfr, a highly pleiotropic developmental gene, was among the most significant additive variants identified. We functionally validated 13 of the additive candidate genes’ (Adgf-A/Adgf-A2/CG32181, bru1, CadN, flapper (CG11073), CG15236, flippy (CG9766), CREG, Dscam4, form3, fry, Lasp/CG9692, Pde6, Snoo), and introduce a novel approach to whole gene significance screens: PEGASUS_flies. Additionally, we identified ppk23, an Acid Sensing Ion Channel (ASIC) homolog, as an important hub for epistatic interactions. We propose a model that suggests genetic modifiers of wing and muscle morphology, nervous system development and function, BMP signaling, sexually dimorphic neural wiring, and gene regulation are all important for the observed differences flight performance in a natural population. Additionally, these results represent a snapshot of the genetic modifiers affecting drop-response flight performance in Drosophila, with implications for other insects.

Keywords:

Animal flight – Drosophila melanogaster – Epistasis – Genetic engineering – Genetic networks – Genetic screens – Genetics – Insect flight

Introduction

Flight is one of the most distinguishing features of many winged insects, especially the taxonomic order Diptera. Colloquially named “flies,” these insects rely on their namesake for many facets of their life history: dispersal, foraging, evasion, migration, and mate finding [1]. Because flight is central to the life history of flies, many of the genes essential for flight are strongly conserved [2, 3]. Flight is an epiphenomenon, comprised of many highly integrated systems working together. As such, flight performance is likely to be a continuously varying trait with a complex genetic basis.

Evaluating the genetic architecture of complex traits is inherently challenging due to the fact that many genes each contribute a small fraction to the overall trait variation [4, 5]. The omnigenic model [6, 7] frames the genetic architecture of complex traits as a network of all expressed genes where variation stems from the action of a very large number of peripheral modifiers that alter the action of core genes and pathways at the center of the network. Genes in the network are connected by edges representing various interaction types (gene-gene, protein-protein, epistatic, etc.). In the case of flight performance, for example, central genes like Wingless [8] and Act88F [9] are essential for wing and indirect flight muscle development, respectively, while peripheral genes would have more subtle effects on flight from systems like metabolism [10], muscle function [11], neuronal function [12, 13], and anatomical development [14, 15]. Peripheral genes are less likely to experience the same degree of purifying selection as central genes, meaning they are more likely to harbor natural variants that can have subtle effects on phenotype.

We can leverage the diversity of natural variants in a population to uncover novel associations between genotype and phenotype, via Genome Wide Association Study (GWAS). The Drosophila Genetics Reference Panel (DGRP) is a panel of 205 inbred and genetically distinct Drosophila melanogaster lines representing a snapshot of natural variation in a population [16, 17]. Previous studies on complex and highly polygenic, quantitative traits have identified many candidate loci contributing to insect- and Drosophila-specific traits [1820], as well as traits affecting human health and disease [2124].

We designed this study to identify genetic modifiers of flight performance and map out a possible network of the underlying genetic architecture. We screened males and females from 197 DGRP lines, then analyzed both sexes, their average, and their difference. We took a multifaceted approach, identifying modifiers at the individual variant (nadditive = 180 variants; nmarginal = 70 variants; nepistatic = 12,161 variants) and network levels (n = 539 genes). We developed a novel application of the human-based PEGASUS program [25] for use with Drosophila and DGRP studies that identified 72 whole genes of significance: PEGASUS_flies. In addition to these findings, we successfully validated 13 candidate genes from the additive approach using Mi{ET1} mutational insertions.

Taken together, our results strongly suggest variation in flight performance across natural populations is affected by non-coding regions of the genome. Many of the genes affecting the genetic architecture for flight performance are known to affect 1) neural development, 2) development of flight musculature, 3) development of wing morphology, and 4) regulation of gene expression. Based on the validation of our candidate genes and strongly significant genes from different analyses, we propose a model to summarize how these genetic modifiers work to affect flight performance.

Results

Variation in flight performance across the DGRP

Cohorts of approximately 80 flies from 197 lines of the DGRP (S1 Table) were tested for flight performance using a flight column [26] (Fig 1A). We recorded high-speed videos for a weak, intermediate, and strong genotype entering the flight column (Fig 1B–1D and File 1 in https://doi.org/10.7910/DVN/ZTDHDC). Based on these videos, we concluded this assay is best for measuring the reaction and response to an abrupt drop.

Fig. 1. DGRP lines show differences in flight performance across lines.
DGRP lines show differences in flight performance across lines.
(A) The flight performance assay measures the average landing height of flies as they fall through a flight column. Vials of flies are sent down the top chute and abruptly stop at the bottom, ejecting flies into a meter-long column. Falling flies will instinctively right themselves and fly to the periphery, doing so at different times (and therefore landing at different heights) depending on their performance ability. (B-D) Collapsed z-stacks of every 10th frame from a high-speed video recorded from the top quarter (0.25m) of the flight column illustrate these performance differences in (B) weak, (C) intermediate, and (D) strong genotypes. (E) Sexual dimorphism exists within genotypes (deviation of red dashed regression line from y = x solid gray line), though sexes are well correlated (r = 0.75, n = 197, P < 1e-36). (F) Sexually dimorphic performances are also apparent in the distribution of mean landing heights for each male (cyan) and female (red) genotype pair (mean ± S.E.M.). Sex-genotype pairs are sorted in order of increasing male mean landing height. Performances for genotypes in B-D are indicated on the distribution with the corresponding color-coded asterisk (*) above the respective genotype position.

There was strong agreement between the mean landing height of males and females for each genotype (r = 0.75; Fig 1E), with males showing higher landing heights than females (male: 0.80m ± 0.06 SD; female: 0.73m ± 0.07 SD; Fig 1F and S2 Table). Since landing height was somewhat sexually dimorphic, we calculated broad sense heritability (H2) separately for each sex (H2Male = 13.5%; H2Female = 14.4%), and confirmed the assay’s repeatability by retesting 12 lines of varied ability, reared 10 generations apart (r = 0.95; S1 Fig). Together, these results point toward genetic (rather than experimental or environmental) variation as a main source of variation in flight between the individual DGRP lines. In addition to analyzing males and females, we also analyzed the average (sex-average) and difference (sex-difference) between sexes (S2 Fig).

Before running the association analysis, we tested whether flight performance was a unique phenotype. We compared our phenotype scores for males and female against publicly available phenotypes on the DGRP2 webserver, as well as visual senescence at three time points [22]. We found no significant regression between flight performance and any of the phenotypes in either sex after correcting for multiple testing (P ≤ 1.67e-3; S3 Table). This result suggests our measure of flight performance is a unique phenotype among those reported.

Association of additive SNPs with flight performance

We conducted a Genome Wide Association Study (GWAS) to identify genetic markers associated with flight performance. We performed an analysis with 1,901,174 common variants (MAF ≥ 0.05) on the additive genetic effects of four sex-based phenotypes: males, females, sex-average, and sex-difference. Some phenotypes covaried with the presence of major inversions (S4 Table), so we analyzed association results using a mixed model (Fig 2A) to account for Wolbachia infection status, presence of inversions, and polygenic relatedness (S3 and S4 Figs), calculated using the DGRP2 webserver.

Fig. 2. Variation in flight performance associated with several additive variants, some of which were functionally validated.
Variation in flight performance associated with several additive variants, some of which were functionally validated.
(A) An additive screen for genetic variants identified several variants that exceeded the traditional [29] DGRP (P ≤ 1e-5) threshold (gray line). Significant variants (red) were spread throughout the genome on all but chromosome 4. Variants for the sex-averaged phenotype are pictured, though other sex-based phenotypes had similar profiles (S5 Fig). (B) Approximately half of all variants were shared with at least one other sex-based analysis, while the other half of all variants was exclusive to a single analysis. (C) Candidate genes were selected from the Bonferroni-corrected variants and those most significant in the sex-average analysis, for which transgenic flies were publicly available. Both sexes were tested for flight performance. Validated genes were determined if there was a significant difference between experimental lines homozygous for an insertional mutant in the candidate gene and their background control lines lacking the insertional mutant (red points, Mann-Whitney-U test, P ≤ 0.05). Very significant candidate genes (CadN, flapper (CG11073), and Dscam4) each had two independent validation lines.

We filtered additive variants with a strict Bonferroni threshold (P ≤ 2.63e-8). Taking a minSNP approach to identify significant genes if their lowest (most significant) variant P-value crossed a threshold [25], we identified six unique variants. Five of these variants mapped to six genes (CG15236, CG34215, Dscam4, Egfr, fd96Ca, Or85d) (Table 1). Variants mapping to Egfr and fd96Ca also mapped to cis-regulatory elements (transcription factor binding sites (TFBS) and a silencer) known to activate during embryogenesis [27, 28]. Of note, Dscam4 was deemed “damaged” in 38 of the lines tested [17]; however, the difference between mean landing heights of flies with the damaged vs. undamaged allele was less than 1 cm (P = 0.32, Welch’s T-test). The allele causing the damaged condition is not the allele associated with flight performance.

Tab. 1. Six additive variants surpassed the Bonferroni significance threshold.
Six additive variants surpassed the Bonferroni significance threshold.

Using the traditional DGRP significance threshold (P ≤ 1e-5) [29], we identified 180 variants across all four sex-based phenotypes (Figs 2B and S5 and S5 Tables). All of the identified variants passed their permutation test (n = 10,000) significance thresholds for each sex-based phenotype, suggesting variants were not associated with flight by random chance. The individual additive variant with the largest effect size mapped to an intron in epidermal growth factor receptor (Egfr; human homolog EGFR) and contributed a 4.5 cm difference in landing height (or 0.97% of the sum of all significant variants) for males and 6.4 cm (1.1% of the sum of all significant variants) for females. For reference, the variant with the smallest significant effect size was 0.017 cm (or 0.0036% of the sum of all significant variants) for males and 0.57 cm (or 0.095% of the sum of all significant variants) for females. Notably, Egfr encodes a key transmembrane tyrosine kinase receptor and was previously identified as a locus influencing wing shape in the DGRP [30, 31]. As a pleiotropic gene, Egfr affects developmental and homeostatic processes throughout the life and anatomy of the fly. It is known for its role in embryonic patterning and has implications in tumorigenesis [32, 33]. When cis-regulatory elements lie in important developmental genes, their effects can magnify as the organism continues through development. These effects can magnify further for genes like Egfr that are receptors known to affect other developmental processes [31]. Accordingly, this variant in Egfr mapped to several overlapping transcription factor binding sites (TFBS) for transcription factors known to affect embryonic development in a highly dose-dependent manner (bcd, da, dl, gt, hb, kni, Med, prd, sna, tll, twi, disco, Trl) [3437], suggesting this variant may play a similar role.

Across the four sex-based analyses, all but 19 variants mapped to intergenic or non-coding regions (putative cis-regulatory regions). Of the non-coding variants, 149 mapped to 136 unique genes (Table 2). These included development and function of the nervous system and neuromuscular junctions, muscle, cuticle and wing morphogenesis, endoplasmic reticulum and Golgi body functions, and regulation of translation. Approximately half of all variants were present in two or three sex-based analyses, while the remainder were unique to one (Fig 2B). Several variants mapped to transcription factors broadly affecting development and neurogenesis [38, 39]. Despite the enrichment for several annotations, we failed to identify any significant gene ontology (GO) categories using GOwinda [40], a GWAS-specific gene set enrichment analysis.

Tab. 2. Aggregated gene and variant counts by sex-based phenotype for each analysis.
Aggregated gene and variant counts by sex-based phenotype for each analysis.

General development and neurodevelopmental genes validated to affect flight performance

We performed functional validations on a subset of the genes mapped from variants identified in the Bonferroni and sex-average analysis. We identified 21 unique candidate genes for which a Minos enhancer trap Mi{ET1} insertional mutation line [41] was publicly available [42] (S1 Table; Adgf-A/Adgf-A2/CG32181, bru1, CadN, CG11073, CG15236, CG9766, CREG, Dscam4, form3, fry, Lasp/CG9692, Pde6, Snoo). Three additional stocks for CadN, Dscam4, and CG11073 were also tested for their strength of association. CREG was also included as a negative control, since no variant exceeded the 25th percentile for significance (P = 0.25).

Candidate genes were functionally validated by comparing the distribution in mean landing heights of stocks homozygous for the insertion and their paired control counterpart (S6 Fig) using a Mann-Whitney-U test (Fig 2C and S6 Table). Several candidate genes were involved in development of the nervous system (CadN, CG9766, CG11073, CG15236, Dscam4, fry, and Snoo) [8, 4351], muscle development (bru1 and Lasp) [11, 52, 53], and transcriptional regulation of gene expression (CREG) [54] Following successful validation of CG9766 and CG11073, two unnamed candidate genes, we are naming them flippy (flip) and flapper (flap) based on the flipping and flapping motions of weaker flies struggling to right themselves in the flight performance assay. For more information on these two genes, see S1 Text for "Putative roles for flippy and flapper".

Association of gene-level significance and interaction networks with flight performance

The minSNP approach on the additive variants prioritizes the identification of genes containing variants with larger effects [25]. However, this approach ignores linkage blocks and gene length, which can bias results. It is important to account for gene length because some genes can be long and exceed 100kb (e.g. neurodevelopmental genes such as CadN, 131kb). One alternative approach is Precise, Efficient Gene Association Score Using SNPs (PEGASUS), which assesses whole gene significance scores by comparing a gene’s variant P-value distributions against a null chi-squared distribution [25]. This approach enriches for whole genes of moderate effect and enables the identification of genes that might go undetected in a minSNP approach.

Because PEGASUS is configured for human populations, we developed PEGASUS_flies, a modified version for Drosophila <https://github.com/ramachandran-lab/PEGASUS_flies>. This platform is configured to work with DGRP data sets, and can be customized to accept other Drosophila/model screening panels. From our additive variants, PEGASUS_flies identified 72 unique genes across all sex-based phenotypes whose gene scores passed a Bonferroni threshold (P ≤ 3.03e-6; Fig 3A and S7 Table). The significant genes were generally different from those identified in the additive minSNP analyses (Figs 3B and S7), though 15 were shared (aru, bves, CG17839, CG32506, CG33110, fry, Gmap, Mbs, mip40, mxt, oys, Pdp1, Rab30, sdk, VAChT). The relatively low overlap between these two gene sets is to be expected, since they prioritize variants of large effect (minSNP) vs. whole genes of moderate effect (PEGASUS_flies). Overall, gene annotations were enriched for neural development and function, wing and general development, Rab GTPase activity, and regulators of transcription. Different sex-based phenotypes varied in how unique certain whole genes were to a given phenotype (Fig 3C). Genes identified in the sex-average analysis were generally shared with the male and female phenotypes, while genes in the sex-difference analysis were typically unique. Interestingly, dsf and sdk were both present in the sex-average and sex-difference, and Ccn was present in both the male and sex-difference. Ccn was also located in the Missouri insertion of chromosome 3R (In_3R_Mo; a significant covariate for male and sex-average analyses), though PEGASUS_flies accounts for linkage blocks so Ccn is still significant.

Fig. 3. PEGASUS_flies identifies different genetic modifiers than the additive screen.
PEGASUS_flies identifies different genetic modifiers than the additive screen.
(A) PEGASUS_flies results plotted as a Manhattan plot. For the sex-average phenotype, several genes (red points, labeled with gene symbol) exceeded a strict Bonferroni significance threshold (gray dashed line, P ≤ 3.43e-6) identified several genes. (B) PEGASUS_flies prioritizes genetic modifiers of moderate effect, taking into account linkage blocks and gene length. Significant PEGASUS_flies (red) compared against genes significant under a minSNP approach for additive variants (blue) have very little overlap between the two sets (purple). (C) Many of the genes PEGASUS_flies identified were unique to a sex-based phenotype, though the sex-average genes were generally found in other analyses.

Taking advantage of the gene-level significance scores, we leveraged publicly available gene-gene and protein-protein interaction networks to identify altered sub-networks of genes that connect to the flight performance phenotype. A local False Discovery Rate (lFDR) was calculated for each sex-based phenotype (S8 Table), for which gene-scores were either–log10 transformed if they passed, or set to 0 if they did not. Transformed scores for each sex-based phenotype were analyzed together in Hierarchical HotNet [55], which returned a consensus network consisting of nine sub-networks of genes (S9 Table). The largest network identified 512 genes and was significantly enriched for several GO terms, including transcription factor binding, histone and chromatin modification, regulation of nervous system development, and regulation of apoptosis (S10 Table). The other eight networks were comprised of 27 genes, which together had several significant GO terms, including regulation of gene expression through alternative splicing, maintenance of the intestinal epithelium, and the Atg1/ULK1 kinase complex (S11 Table).

Association of epistatic interactions with flight performance

Epistatic interactions account for a substantial fraction of genetic variation in complex traits [56], but they are computationally and statistically challenging to identify. To circumvent the barriers associated with an exhaustive, pairwise search (n = 1.81E12), we focused our search area with MArginal ePIstasis Test (MAPIT). MAPIT is a linear mixed modeling approach that identifies variants more likely to have an effect on other variants. These putative hub variants represent more central and interconnected genes in a larger genetic network proposed by the omnigenic model [6, 7]. Accordingly, we identified 70 unique, significant marginal variants exceeding a Bonferroni threshold (P ≤ 2.56e-8) across male, female, and sex-average phenotypes, but none in the sex-difference analysis (S8 Fig and S12 Table). We tested these 70 marginal variants for pairwise interactions against all other SNPs in the data set and found 20 unique marginal variants with significant pairwise interactions that passed the Bonferroni threshold (S13 Table). Some of these interactions were between genes containing marginal variants, but not necessarily between the marginal variants themselves (Fig 4A). For example, ppk23 and sog both contain significant marginal variants but the significant pairwise interactions were between marginal and non-marginal variants. This highlights an important benefit of using several approaches for finding different types of additive, marginal, or epistatic effects within the same gene. Additionally, since marginal variants represent those more likely to interact with other variants, their interaction between genes containing significant marginal variants suggests a highly interconnected genetic architecture underlying flight performance. The breadth of epistatic interactions from a small, focused subset of marginal variants supports an important role for epistasis in the genetic architecture of flight performance. There are likely many more variants that interact with one another beyond the limited subset of 70 marginal variants we tested. We briefly recapitulate these findings in order of the male, female, and sex-average results, though a more comprehensive survey is available in S1 Text under "Association of epistatic interactions with flight performance, continued".

Fig. 4. Flight performance is a larger complex trait comprised of several smaller traits.
Flight performance is a larger complex trait comprised of several smaller traits.
(A) The genetic architecture of epistatically interacting genes shared ppk23 as a more central node. (B) Whole genes and minSNP genes were not identified in more than three analyses, while roughly half or more genes were unique to each analysis. (C) Flight performance has a complex genetic architecture, with the key developmental gene Egfr and BMP signaling pathway contributing to wing and neurodevelopment. These processes are both important for structuring the sensory organs that enable the fly to use mechanosensory channels for proprioception. Signals from the sensory organs on the wing, head, and body travel to the brain and thoracic ganglion, which sends signals through the motor neurons to the direct and indirect flight musculature that is also differentially assembled and innervated to generate power and control the wing angle during flight.

In males, there were seven significant marginal variants that mapped to five genes (CG5645, CG18507, cv-c, sog, Ten-a). Of the variants, only the one (X_15527230_SNP) that mapped to a novel transcription start site in the BMP antagonist of short gastrulation (sog; human ortholog of CHRD) had significant interactions. This marginal SNP interacted with 42 other variants across 28 unique genes (S13 Table). A quarter of these genes are important in neuron development, signaling, and function.

In females, there were 14 significant marginal variants that mapped to several intergenic variants and six genes (CG6123, CG7573, CG42741, ppk23, Src64B, twi). Most notably, four variants mapped to a 1,002 bp region downstream of pickpocket 23 (ppk23; human homologs in ASIC gene family) and accounted for the majority of epistatic interactions in the female analysis. ppk23 is a member of the degenerin (DEG)/epithelial Na+ channel (ENaC) gene family that functions as subunits of non-voltage gated, amiloride-sensitive cation channels. It is involved in chemo- and mechanosensation, typically in the context of foraging, pheromone detection, and courtship behaviors [57, 58].

In the sex-average analysis, there were 62 significant marginal variants that mapped to intergenic regions and 21 genes (Art2, CG10936, CG15630, CG15651, CG18507, CG3921, CG42671, CG42741, CG5645, CG6123, CG9313, CR44176, cv-c, Fad2, natalisin, ppk23, Rbfox1, Rgk1, Src64B, twi; Fig 4A). Of the 62 marginal variants, 18 had significant epistatic interactions with other variants in the genome, with the majority in intergenic regions (n = 9) and ppk23 (n = 7), while individual variants mapped to CG42671, CG10936, CG9313, and CG15651 (S13 Table). Again, ppk23 had the greatest number of epistatic interactions of a single gene, and many of these interactions were with variants that mapped to genes with significant marginal variants themselves (A2bp1, cv-c, Fad2, CG9313, CG10936, CG42741, Rgk1, sog, Src64B, twi, Ten-a). Notably, ppk23’s mapped interactions were with genes that collectively had significant GO term enrichment for neuronal growth, organization and differentiation (S14 Table). There were two other groups of variants: one group in CG42671 and the other containing six intergenic variants in a 669 bp region (chr3L:6890373–6891042), that also had an abundance of epistatic interactions with significant enrichment for several GO categories related to development and function of the nervous system (S15 and S16 Tables).

No evidence for adult transcriptome variation affecting flight performance

Since many variants mapped to cis- and trans-regulatory genes, we sought to test whether regulatory variation was affecting developmental or adult homeostasis. Accordingly, we performed a Weighted Gene Co-expression Network Analysis (WGCNA) [59] using 177 publicly available DGRP transcriptomic profiles for young adults of both sexes [60]. We clustered genes by similarity in expression profile, then correlated those clusters’ eigenvalues with the mean and standard deviation of flight performance, as well as the proportion of flies that fell through the column versus the total assayed. No clusters across sex or phenotype had a significant correlation. This result supports our previous observation that many of the significant variants map to genes involved in pre-adult development, rather than genes that are likely to have variable expression levels as adults assayed under presumably homeostatic conditions [60] (S9 Fig). Accordingly, we recommend future studies on complex traits should explore gene expression or similar phenotypes targeted at relevant developmental stages, rather than only during later or adult stages when a phenotype is measured.

Flight performance is modulated by an interconnected genetic architecture

The genetic architecture of flight performance is comprised of many different types of genetic modifiers. Many of the variants map to genes that are found across analytic platforms (Fig 4B). Most variants were unique to a single analysis, suggesting that association studies should consider using multiple different analyses to enhance the power to detect variants and genes in their study. However, many genes were identified in two (148) or three (23) analyses. Those involved in three analyses include: aru, CG2964, CG13506, CG15651, CG17839, CG42671, CycE, daw, Diap1, Egfr, fz2, Gart, Gmap, Mbs, MED23, mip40, mxt, Pdp1, Rab30, rhea, sog, sona, Tgi. This suggests that individual genes can contain variants with different types of effects or have differential contributions to the overall genetic architecture. A complete lookup table of all genes and genes identified from variants is available in S17 Table.

Discussion

We tested flight performance of 197 DGRP lines, identifying several additive and marginal variants, epistatic interactions, whole genes, and a consensus network of altered sub-networks that associated with variation in our phenotype. Many putative cis-regulatory variants mapped to genes with annotations for wing morphology, indirect flight muscle performance, and development of sensory neurons and neuromuscular junctions. We demonstrate that implementation of complementary approaches can expand the breadth of genetic modifiers identified and improve genomic predictions in mapping the genotype-phenotype landscape [29]. Our study corroborates this observation and lays out four additional computational approaches that can expand on the traditional minSNP output from the DGRP2 webserver. These results expand our understanding of the genetic architecture of complex traits, since they provide greater context at the whole gene (PEGASUS_flies) and interaction/network levels (MAPIT, Hierarchical HotNet, PLINK’s–epistasis feature). Associating genes from individual variants can be tricky [25], and finding epistatic interactors can be computationally and statistically demanding [61]. Furthermore, by combining these approaches, we add to a growing body of literature emphasizing the importance of a multi-faceted approach in elucidating the genetic architecture of complex traits.

Neurodevelopmental genes play an important role in modifying flight performance

There were a number of genes associated with the development of the nervous system across all analyses, including a notable overlap between the additive minSNP and whole gene screens (aru, ChAT, Ccn, DIP-δ, dsf, dsx, fry, Mbs, sdk, VAChT). Successful validation of known genes involved in neurodevelopment (CadN, Dscam4, fry, Snoo), confirmed its importance. Specifically, these four genes are known to work together for developing and patterning the small, sensory, hair-like structures that line the fly’s body and wings (microchaete). These four genes are also known to facilitate innervation of these structures and connect them to the central nervous system (CNS) through type IV dendritic arborization sensory neurons [48, 6266]. Interestingly, we identified 41 pickpocket, olfactory receptor neuron (ORN), gustatory receptor neuron (GRN), and ionotropic receptor (IR) genes that aid in signal reception on microchaete. These important mechano- and chemosensory structures suggest that they might play an important role in flight performance. Of note, only six of these 41 genes were previously identified from an olfactory screen with 14 separate odors in a past DGRP GWA study (Gr59d, Ir41a, Ir60d, Or24a, ppk10 and ppk12) [18], suggesting a putative role for these genes in flight performance and a potential explanation for the identification of Or85d as a Bonferroni-corrected additive variant in the minSNP analysis.

Interestingly, pickpocket 23 (ppk23) was identified as a central node in the marginal and epistasis analyses. Pickpocket family genes are a conserved group of acid sensing ion channels (ASIC family in humans) that are known to work with chemosensory receptors for pheromone detection and are often studied in the context of courtship [57, 67], as well as roles in proprioception and mechanotransduction along peripheral sensory neurons [58, 68, 69]. Accordingly, we hypothesize a proprioceptive role for ppk23 during flight. ppk23 ’s breadth of epistatic interactions suggests it plays a more central role in flight performance than previously expected. This includes epistatic interactions with genes identified in previous flight performance studies that were not found in our other analyses (cac, Hk, Sh, and slo) [7073]. Pickpocket genes, including ppk23, may also play an important role in the central pattern generator (CPG) circuits of the thoracic ganglion, which acts as an important neuro-regulator of flight [74]. These circuits are responsible for ’fictive’ behavioral patterns (e.g. walking, flying, preening), or repetitive behaviors that can be maintained in the absence of sensory inputs. Fictive behaviors, such as the flight control phenotype we measured, can incorporate proprioceptive cues to modulate neuromuscular outputs that affect the organism’s behavior. Mutants of the pickpocket gene ppk1 are known to affect larval locomotion by disrupting the timing of the output signals [75, 76]. As members of the degenerin/epithelial sodium channel (DEG/ENaC) family, pickpocket family genes may be important for promoting presynaptic homeostatic plasticity. Here, pickpocket genes create sodium ion "leaks" that depolarize the presynaptic membrane and promote calcium influx to aid in stabilizing neuronal function following the postsynaptic release of neurotransmitters [77, 78]. These pickpocket genes dimerize into larger pickpocket subunit assemblies, creating channels with different and unique physiological properties [77]. Thus, pickpocket genes, including ppk23, may also be important for ensuring motor neurons in flight musculature are functional throughout flight.

There was another noteworthy epistatic interaction between ppk23 and fruitless (fru), an important transcription factor involved in differentiating sex-specific neural circuits that affects fitness-related behaviors, like courtship [79]. fru is also a modulator of CPG activity during flight [80], and is known to co-localize with ppk23 in the thoracic ganglion [4, 58, 67, 81, 82]. Both also work with doublesex (dsx), another transcription factor that affects sex-specific neural circuits and was found in the whole gene, sex-difference analysis [8385]. These genes suggest a potential genetic mechanism underlying the observed sexual dimorphism in flight performance (see S1 Text for more information on "Potential genetic sources of sexual dimorphism in flight performance").

Natural variants in flight muscle genes associate with variation in flight performance

Two muscle-associated genes with known roles in flight were identified and validated from the additive screen. Lasp (human ortholog LASP1) modifies sarcomere and thin filament length, and myofibril diameter [52]; bruno 1 (bru1 or aret; human homolog CLEF1 and CLEF2) is a transcription factor that controls alternative splicing of myofibrils in the indirect flight muscle [11, 53], among other developmental processes.

Interestingly, we also identified two genes affecting muscle function through PEGASUS_flies that were previously validated in the literature. Tropomodulin (tmod; human homolog TMOD1) is responsible for muscle function and Glycerol-3-phosphate dehydrogenase 1 (Gpdh1; human homolog GPD1) affects metabolism within muscles [10, 86]. Neither gene contained a significant variant exceeding the additive screen’s significance threshold (P ≤ 1e-5), demonstrating PEGASUS_flies’ ability to identify genetic modifiers overlooked in a traditional minSNP approach.

Variation in wing development contributes to variation in flight performance

One of the additive Bonferroni-corrected variants mapped to Egfr, a canonical developmental gene known to harbor natural variants that can modify wing morphology and flight performance [30, 31]. Egfr is pleiotropic, though one of the main roles it plays is in conjunction with the Bone Morphogenetic Protein (BMP) signaling pathway [13, 31, 87], an established pathway known to affect wing development. BMP signaling forms dose-dependent gradients influencing wing size, shape, and venation patterning [8891]. It can also affect sensory and neuromuscular circuits in flight structures [8, 92]. In addition to our identification of several modifiers of BMP signaling across all analyses (cmpy, Cul2, cv-2, cv-c, dally, daw, dpp, egr, gbb, hiw, kek5, Lis-1, Lpt, lqf, ltl, Mad, nmo, scw, Snoo, sog, srw, tkv, trio), we functionally validated Sno Oncogene (Snoo; human homolog SKI), an important component of the BMP signaling pathway that affects dendritic tiling and morphogenesis, wing shape, and the development of sensory organs on the wing (e.g. microchaete and campaniform sensilla) [8, 50, 93, 94]. We also identified a significant marginal variant coding for a novel transcriptional start site in short gastrulation (sog; human homolog Chordin), a dpp antagonist in patterning the dorsoventral axis of the wings [89, 95, 96]. Additionally, sog is a known source of natural variants that modify flight performance in natural populations [13]. This particular sog variant had epistatic interactions in other genes containing marginal variants, including ppk23 and CG42671, suggesting a more interconnected role for this antagonist of BMP signaling in modifying flight performance.

Variation in gene regulation drives variation in flight performance

Similar to other DGRP studies [19, 29, 97], the majority of significant variants we had the power to test mapped to non-coding or intergenic regions, which are hypothesized to associate with cis-regulatory elements [29]. Under the omnigenic model, variation in these elements may have a disproportionate effect downstream when they occur in trans-regulatory genes (i.e. transcription factors, splicosomal proteins, and chromatin modifiers), since they themselves modify the regulation of other genes [32, 33]. One additive variant that passed the Bonferroni threshold mapped to an intronic region in forkhead domain 96Ca (fd96Ca; human homologs FOXB1 and FOXB2), a dorsal (dl) transcription factor binding site (TFBS), and a silencer for histone deacetylase 1 (HDAC1). fd96Ca is a forkhead box transcription factor expressed in neuroblasts along the longitudinal axis of the embryo and in some sensory neurons in the embryonic head [98]. Mis-regulation of transregulatory genes in developmental tissues (e.g. neuroblasts and the brain) may subsequently affect the ontogeny of tissues derived from it downstream. Another example highlighting the importance of trans-regulatory genes lies in the transcription factor CREG. It was initially selected as a negative control in the validation screen, though we demonstrate it has a significant effect on flight ability in our validation screen of candidate genes. The significant effect of the insert we tested in the screen may be a result of a difference between the insertional mutation and variation in the SNPs we have the power to test.

Further evidence supporting the role of regulatory elements affecting flight performance lies more broadly in our validation screen. The Mi{ET1} constructs introduced genetic variation into non-coding regions of genes, but not necessarily near the significant variants identified in our GWA screen. Thus, genetic variation across a gene, rather than at a single variant, is capable of affecting phenotypic variation. Our successful validation for 13/21 of the unique genes tested is near the approximately 70% success rate of other DGRP studies’ validation screens [29]. Interestingly, the insertion of the constructs into intronic regions both positively and negatively affected performance, even when done at independent sites in the same gene (CadN and Dscam4; see Fig 2C). Additionally, some candidate genes only validated in one sex, despite being identified in the sex-average analysis. These findings suggesting a more nuanced impact of genetic variation in cis-regulatory regions that can have differential sex-effects, possibly resulting from sexually-dimorphic epistatic interactions. Overall, their validation supports an important role for non-coding regions and regulation of gene expression, as well as the genes the constructs were inserted in, as modifiers of flight performance.

A proposed model for understanding the genetic architecture of flight performance

Flight performance is likely an epiphenomenon of several interconnected complex traits. While we are unable to identify and validate every modifier, we likely identified the main features of the genetic architecture. Our results are biased, in part, by the annotations available for many of these genes. Many genes involved in the neural system are studied in the context of development, rather than their role in the adult stage. As a result, there is often relatively sparse information on their function in the adult fly. Our data suggests a means for studying these variants further in adults. Accordingly, we propose the following model to synthesize our findings primarily based on strongly significant genes we identified and validated, and supported by other genes we identified in the network and epistasis analyses (Fig 4C).

Epidermal growth factor receptor is a key gene in a canonical developmental pathway. It is a pleiotropic gene affecting wing morphology, sensory organ development, and neurodevelopment, on its own and through the BMP signaling pathway [3133]. Proper development of these structures and circuits enables a well-connected peripheral nervous system to receive external, proprioceptive stimuli (through CadN, Dscam4, fry, ppk23, Snoo) and may also help shape the development of central pattern generator (CPG) circuits in the thoracic ganglion. As mentioned above, CPG circuits produce repetitive behavioral movements that may be modulated in either a sex-specific manner through ppk23, fru, and dsx, or a non-sex-specific manner with many other genes, including ppk23 and other pickpocket family genes. The brain and thoracic ganglion process proprioceptive signals and activate motor neurons innervating the direct (control) and indirect (power) flight musculature (through bru1, Gpdh1, Lasp, tmod) at neuromuscular junctions. Activation of these muscles allows the properly developed wings to both flap and flip and generate lift [99]. The model proposed may help generate testable hypotheses that could help elucidate the architecture of flight as a complex trait.

Conclusions

We highlight the benefits of conducting multiple computational analyses in evaluating variants, genes, and interaction networks and their contribution to the genetic architecture of flight performance by conducting a GWAS with the Drosophila Genetic Reference Panel lines. We also introduce PEGASUS_flies, an open-source and novel approach to whole gene significance screening. We present evidence that genetic modifiers with annotations for neurodevelopment, muscle and wing development, and regulation of other genes affect flight performance. We also uncover an important role for ppk23 as an important epistatic hub in modulating flight performance, raising the importance of mechanosensation and proprioception in helping flies react and respond to an abrupt drop.

Materials and methods

Drosophila stocks and husbandry

All stocks were obtained from Bloomington Drosophila Stock Center (https://bdsc.indiana.edu/), including 197 Drosophila Genetic Reference Panel (DGRP) lines [17], 23 Drosophila Gene Disruption Project lines using the Mi{ET1} construct [41, 100], and two genetic background lines (w1118 and y1w67c23; S1 Table).

Flies were reared at 25° under a 12-h light-dark cycle. Stocks were density controlled and grown on a standard cornmeal media [101]. Two to three days post-eclosion, flies were sorted by sex under light CO2 anesthesia and given five days to recover before phenotyping.

Flight performance assay

Flight performance was measured following the protocol refined by Babcock and Ganetzky [26]. Briefly, each sex-genotype combination consisted of approximately 100 flies, divided into groups of 20 flies across five glass Drosophila culture vials. These vials were gently tapped to draw flies down, and unplugged before a rapid inversion down a 25 cm chute. Vials stopped at the bottom, ejecting the flies into a 100 cm long x 13.5 cm diameter cylinder lined with a removable acrylic sheet coated in TangleTrap adhesive. Free falling flies instinctively right themselves before finding a place to land, which ended up immobilizing them at their respective landing height. Flies that passed through the column were caught in a pan of mineral oil and were excluded from the analysis.

After all vials in a run were released, the acrylic sheet was removed and pinned to a white poster board. A digital image was recorded on a fixed Raspberry PiCamera (V2) and the x,y coordinates of all flies were located with the ImageJ/FIJI Find Maxima function, set with a light background and noise tolerance of 30 [102]. For each sex-genotype combination, the mean landing height was calculated for only the flies that landed on the acrylic sheet.

High-speed video capture of flight column

High-speed videos of flies leaving the flight column were recorded at 1540 frames per second using a Phantom Miro m340 camera recording at a resolution of 1920 x 1080 with an exposure of 150 μs (File 1 in https://doi.org/10.7910/DVN/ZTDHDC). The camera was equipped with a Nikon Micro NIKKOR (105 mm, 1:2.8D) lens and Veritas Constellation 120 light source.

Estimating heritability

Individual fly landing heights were adjusted for the presence of inversions and Wolbachia status by sex and genotype, as calculated by the DGRP2 webserver. Using these adjusted landing heights by sex, we performed a random effects analysis of variance using the R (v.3.5.2) package lme4 (v.1.1.23): Y ~ μ + L + ε. Here, Y is the adjusted flight score, μ is the combined mean, L is the line mean, and ε is the residual. From this, sex-specific broad sense heritability (H2) estimates were calculated from the among line (σL2) and error (σE2) variance components: H2 = σL2 / (σL2 + σE2).

Genome wide association mapping

Flight performance scores for males and females were submitted to the DGRP2 GWAS pipeline (http://dgrp2.gnets.ncsu.edu/) [16, 17] and results for each sex, and the average (sex-average) and difference (sex-difference) between them were all considered (S2 Table). In total, 1,901,174 variants with a minor allele frequency (MAF) ≥ 0.05 were analyzed (File 2 in https://doi.org/10.7910/DVN/ZTDHDC). All reported additive variant P-values result from a linear mixed model analysis, including Wolbachia infection and presence of five major inversions as covariates. Variants were filtered for significance using the conventional P ≤ 1e-5 threshold [29]. Effect size estimates were calculated as one-half the difference between the mean landing heights for lines homozygous for the major vs. minor allele. The contribution of individual variants to the overall effects was estimated as the absolute value of an individual variant’s effect size divided by the sum of the absolute values for all conventionally significant (P ≤ 1e-5) variants’ effect sizes.

Candidate gene disruption screen

Candidate genes were validated using insertional mutant stocks generated from Gene Disruption Project [42]. These stocks contain a Minos enhancer trap construct Mi{ET1} [41] and were built on either w1118 or y1 w67c23 backgrounds (BDSC_6326 and BDSC_6599, respectively).

Control and experiment line genetic backgrounds were isogenized with five successive rounds of backcrossing the insertional mutant line to its respective control. Validation of flight phenotypes was done using offspring of single-pair (1M x 1F) crosses between the control and insert lines. Heterozygous flies from these crosses were mated in pairs and the homozygous offspring lacking the insertion were collected as the control. Candidate heterozygous/homozygous positive lines were mated as pairs once more and lines producing only homozygous positive offspring were used as experimental lines (S1 Fig). Experimental lines were checked for a GFP reporter three generations later to confirm their genotype. The finalized recombinant backcrossed control and experimental lines for each sex-genotype combination were assayed for flight performance, and tested for significance, via Mann-Whitney U-tests.

Calculating gene-score significance

Gene-scores were calculated using Precise, Efficient Gene Association Score Using SNPs (PEGASUS) [25]. Originally implemented with human datasets, we modified the program to work with Drosophila datasets, which we call PEGASUS_flies. It also contains default values adjusted for Drosophila, a linkage disequilibrium file, and gene annotations drawn from the FB5.57 annotation file, available on the DGRP webserver. PEGASUS_flies is available at: https://github.com/ramachandran-lab/PEGASUS_flies, and as File 4 in https://doi.org/10.7910/DVN/ZTDHDC.

Identifying altered sub-networks of gene-gene and protein-protein interaction networks

Returned gene-scores were filtered for genes of high confidence using the Twilight package (v.1.60.0) in R (Scheid and Spang 2005). Here, we estimated the local False Discovery Rate (lFDR) of all previously output gene scores using the twilight function. Taking the inflection point of the (1—lFDR) curve, our high-confidence gene scores ranged from 0.65–0.73 for the four, sex-based phenotypes (S8 Table). High confidence genes were–log10 transformed, while the remaining were set to 0.

Hierarchical HotNet was used to identify altered sub-networks of interacting genes or proteins [55] based on network topology generated from several gene-gene or protein-protein interaction networks. The four adjusted, sex-based gene-score vectors were mapped in the program to fifteen interaction networks obtained from High-quality INTeractomes (HINT) [103], the Drosophila Interactions Database (Droidb) [104, 105], and the Drosophila RNAi Screening Center (DRSC) Integrative Ortholog Prediction Tool (DIOPT) [106]. Consensus networks were calculated from 100 permutations of all four gene-score vectors on each of the fifteen interaction networks and filtered to include at least three members. The largest sub-network and the remaining eight sub-networks were passed to Gene Ontology enRIchment anaLysis and visuaLizAtion tool (GOrilla) to identify enrichment for gene ontology (GO) categories [107, 108].

Screening for epistatic interactions

Epistatic hub genes were identified using MArginal ePIstasis Test (MAPIT), a linear mixed modeling approach that tests the significance of each SNP’s marginal effect on a chosen phenotype. MAPIT requires a complete genotype matrix, without missing data. SNPs were imputed using BEAGLE 4.1 [109, 110] and then filtered for MAF ≥ 0.05 using VCFtools (v.0.1.16) [111]. MAPIT was run using the Davies method on the imputed genome (File 2 in https://doi.org/10.7910/DVN/ZTDHDC), DGRP2 webserver-adjusted phenotype scores for each sex-based phenotype (S2 Table), DGRP2 relatedness matrix and covariate file containing Wolbachia infection and the presence of five major inversions [17].

Resulting marginal effect P-values (File 3 in https://doi.org/10.7910/DVN/ZTDHDC) were filtered to a Bonferroni threshold (P ≤ 2.56e-8) and tested for pairwise epistatic interactions in a set-by-all framework against the initial 1,901,174 SNPs (unimputed; MAF ≥ 0.05) using the PLINK–epistasis flag (v.1.90) [112]. Results were filtered for all P-values that exceeded a Bonferroni threshold, calculated as 0.05 / (the number of Bonferroni marginal effect P-values x 1,901,174 SNPs).

Annotating FBgn and orthologs

Annotations and unreferenced descriptors of gene functions, expression profiles, and orthologs were gathered from autogenerated summaries on FlyBase [38, 39]. These summaries and descriptors were compiled from data supplied by the Gene Ontology Consortium [27, 28], the Berkeley Drosophila Genome Project [113], FlyAtlas [114], The Alliance of Genome Resources Consortium [115], modENCODE [38], PAINT [116], the DRSC Integrative Ortholog Prediction Tool (DIOPT) [106], and several transcriptomics and proteomic datasets [11, 12, 43, 114, 117119].

Flybase gene (FBgn) identifiers were converted to their respective D. melanogaster (Dmel) or H. sapiens (Hsap) gene symbols using the Drosophila RNAi Stock Center (DRSC) Integrative Ortholog Prediction Tool (DIOPT) [106]. FBgn were filtered for all high to moderate confidence genes, or low confidence genes if they contained the best forward and reverse score.

Calculating an empirically simulated significance threshold

In order to identify any spurious associations between the DGRP genotyped lines and the four phenotypes of interest, we performed a permutation-based test using the—mperm flag in PLINK v.1.90. For each phenotype, we performed 10,000 permutations of the phenotype values across lines and tested those randomly assigned values for an association to the permuted phenotype. We find that all of the variants associated with the four phenotypes in the standard GWAS framework remained significant after filtering based on the permutation p-value (P < 0.05).

GO term analysis

GOWINDA [40] was implemented to perform a Gene Ontology (GO) analysis that corrects for gene size in GWA studies. We conducted this analysis for male (n = 418), female (n = 473), sex-average (n = 527), and sex-difference (n = 214) candidate SNPs exceeding a relaxed P < 1e-4 significance threshold, against the 1,901,174 SNPs with MAF ≥ 0.05. We ran 100,000 simulations of GOWINDA using the gene mode and including all SNPs within 2000 bp.

Gene Ontology enRIchment anaLysis and visuaLizAtion tool (GOrilla) [107, 108] was run on PEGASUS_flies gene-scores and Hierarchical Hotnet sub-networks using the default commands and a gene list compiled from all genes available in the FB5.57 annotation file.

Weighted Gene Co-expression Network Analysis

To test whether ambient adult transcriptomes could explain the observed phenotypic variation, we turned to the publicly available DGRP2 microarray data, downloaded from the DGRP2 webserver [17]. These data represent the transcriptomes for untreated young adult flies of each sex. We performed Weighted Gene Co-expression Network Analysis (WGCNA) analyses using the WGCNA R package [59] to cluster and correlate the expression profiles of genes from 177 shared, DGRP lines. This analysis was run using the following parameters: power = 16 (from soft threshold analysis ≥ 0.9), merging threshold = 0.0, signed network type, maximum blocksize = 1000, minimum module size = 30.

Supporting information

S1 Fig [tif]
DGRP lines’ mean flight performance is highly repeatable across generations.

S2 Fig [m]
Sex-average and sex-difference phenotypic distributions are amenable to an association study.

S3 Fig [a]
QQ-plots show enrichment for some additive variants across each of the sex-based phenotypes.

S4 Fig [tif]
Top additive associations are spaced throughout the genome.

S5 Fig [a]
Additional sex-based phenotype Manhattan plots for additive analysis.

S6 Fig [tif]
Genetic crosses performed for deriving experimental and control stocks used to validate candidate genes.

S7 Fig [a]
Significant whole genes are distributed throughout the genome and sex-based phenotypes.

S8 Fig [a]
Significant marginal variants are unevenly distributed across sex-based phenotypes.

S9 Fig [wgcna]
Trait-relationship correlation matrix shows no correlation between measured phenotypes and young adult transcriptome.

S1 Table [xlsx]
stocks used in this study.

S2 Table [xlsx]
Raw and adjusted flight performance metrics.

S3 Table [xlsx]
No significant correlations were observed between flight performance and other DGRP phenotypes.

S4 Table [xlsx]
Up to two inversions were significant covariates in three of the sex-based analyses.

S5 Table [xlsx]
Several additive variants associated with flight performance.

S6 Table [xlsx]
Several candidate genes were validated for flight performance.

S7 Table [xlsx]
Several gene-scores pass a Bonferroni threshold across all four sex-based phenotypes.

S8 Table [xlsx]
Twilight-estimated local False Discovery Rate (lFDR) cutoff thresholds for gene-scores.

S9 Table [xlsx]
sub-network gene annotations.

S10 Table [xlsx]
Large sub-network from is enriched for trans-regulatory factors and neurodevelopmental Gene Ontology terms.

S11 Table [xlsx]
Collection of smaller sub-networks from are collectively enriched for mRNA splicing and autophagy Gene Ontology terms.

S12 Table [xlsx]
Significant marginal variants identified from MAginal ePIstasis Test ().

S13 Table [xlsx]
Epistatic interactions play a large role in shaping the genetic architecture of flight performance.

S14 Table [xlsx]
Epistatic interactions with () are enriched in a Gene Ontology (GO) term analysis form neurodevelopmental genes.

S15 Table [xlsx]
Genes mapped to from epistatic interactions with are significantly enriched for neurodevelopment in a Gene Ontology (GO) analysis.

S16 Table [xlsx]
Gene set enrichment analysis for significant epistatic interactors within a 669 bp intergenic region between chr3L:6890373–6891042 suggests enrichment for neurodevelopmental Gene Ontology categories.

S17 Table [xlsx]
Master lookup table for all genes identified.

S1 Text [docx]
Supplemental results.


Zdroje

1. Brodsky AK. The evolution of insect flight: Oxford University Press; 1994.

2. Edwards JS. The evolution of insect flight: Implications for the evolution of the nervous system. Brain Behavior and Evolution. 1997;50(1):8–12. doi: 10.1159/000113317 WOS:A1997XE45100002. 9209762

3. Ugur B, Chen KC, Bellen HJ. Drosophila tools and assays for the study of human diseases. Disease Models & Mechanisms. 2016;9(3):235–44. doi: 10.1242/dmm.023762 WOS:000371439600003. 26935102

4. Pavlou HJ, Goodwin SF. Courtship behavior in Drosophila melanogaster: towards a ’courtship connectome’. Current Opinion in Neurobiology. 2013;23(1):76–83. doi: 10.1016/j.conb.2012.09.002 WOS:000314562900013. 23021897

5. Weitkunat M, Schnorrer F. A guide to study Drosophila muscle biology. Methods. 2014;68(1):2–14. doi: 10.1016/j.ymeth.2014.02.037 24625467

6. Liu XY, Li YI, Pritchard JK. Trans Effects on Gene Expression Can Drive Omnigenic Inheritance. Cell. 2019;177(4):1022–+. doi: 10.1016/j.cell.2019.04.014 WOS:000466843000020. 31051098

7. Boyle EA, Li YI, Pritchard JK. An Expanded View of Complex Traits: From Polygenic to Omnigenic. Cell. 2017;169(7):1177–86. doi: 10.1016/j.cell.2017.05.038 WOS:000403332400008. 28622505

8. Quijano JC, Stinchfield MJ, Zerlanko B, Gibbens YY, Takaesu NT, Hyman-Walsh C, et al. The Sno Oncogene Antagonizes Wingless Signaling during Wing Development in Drosophila. Plos One. 2010;5(7). doi: 10.1371/journal.pone.0011619 WOS:000279980800008. 20661280

9. Nongthomba U, Pasalodos-Sanchez S, Clark S, Clayton JD, Sparrow JC. Expression and function of the Drosophila ACT88F actin isoform is not restricted to the indirect flight muscles. Journal of muscle research and cell motility. 2001;22(2):111–9. doi: 10.1023/a:1010308326890 11519734

10. Montooth KL, Marden JH, Clark AG. Mapping determinants of variation in energy metabolism, respiration and flight in Drosophila. Genetics. 2003;165(2):623–35. 14573475

11. Kao SY, Nikonova E, Ravichandran K, Spletter ML. Dissection of Drosophila melanogaster Flight Muscles for Omics Approaches. Jove-Journal of Visualized Experiments. 2019;(152). doi: 10.3791/60309 WOS:000493379500080. 31680668

12. Okada H, Ebhardt HA, Vonesch SC, Aebersold R, Hafen E. Proteome-wide association studies identify biochemical modules associated with a wing-size phenotype in Drosophila melanogaster. Nature Communications. 2016;7. doi: 10.1038/ncomms12649 WOS:000384969200001. 27582081

13. Marcus JM. The development and evolution of crossveins in insect wings. Journal of Anatomy. 2001;199:211–6. doi: 10.1046/j.1469-7580.2001.19910211.x WOS:000170738500022. 11523825

14. Lobell AS, Kaspari RR, Negron YLS, Harbison ST. The Genetic Architecture of Ovariole Number in Drosophila melanogaster: Genes with Major, Quantitative, and Pleiotropic Effects. G3-Genes Genomes Genetics. 2017;7(7):2391–403. doi: 10.1534/g3.117.042390 WOS:000404991600035. 28550012

15. Watanabe K, Stringer S, Frei O, Mirkov MU, de Leeuw C, Polderman TJC, et al. A global overview of pleiotropy and genetic architecture in complex traits. Nature Genetics. 2019;51(9):1339–+. doi: 10.1038/s41588-019-0481-0 WOS:000484010800010. 31427789

16. Mackay TFC, Richards S, Stone EA, Barbadilla A, Ayroles JF, Zhu DH, et al. The Drosophila melanogaster Genetic Reference Panel. Nature. 2012;482(7384):173–8. doi: 10.1038/nature10811 WOS:000299994100029. 22318601

17. Huang W, Massouras A, Inoue Y, Peiffer J, Ramia M, Tarone AM, et al. Natural variation in genome architecture among 205 Drosophila melanogaster Genetic Reference Panel lines. Genome Research. 2014;24(7):1193–208. doi: 10.1101/gr.171546.113 WOS:000338185000013. 24714809

18. Arya GH, Magwire MM, Huang W, Serrano-Negron YL, Mackay TFC, Anholt RRH. The Genetic Basis for Variation in Olfactory Behavior in Drosophila melanogaster. Chemical Senses. 2015;40(4):233–43. doi: 10.1093/chemse/bjv001 WOS:000355703500004. 25687947

19. Chow CY, Wolfner MF, Clark AG. Large Neurological Component to Genetic Differences Underlying Biased Sperm Use in Drosophila. Genetics. 2013;193(1):177–85. doi: 10.1534/genetics.112.146357 WOS:000312894700011. 23105014

20. Battlay P, Leblanc PB, Green L, Garud NR, Schmidt JM, Fournier-Level A, et al. Structural Variants and Selective Sweep Foci Contribute to Insecticide Resistance in the Drosophila Genetic Reference Panel. G3-Genes Genomes Genetics. 2018;8(11):3489–97. doi: 10.1534/g3.118.200619 WOS:000449381500009. 30190421

21. Chow CY, Kelsey KJP, Wolfner MF, Clark AG. Candidate genetic modifiers of retinitis pigmentosa identified by exploiting natural variation in Drosophila. Human Molecular Genetics. 2016;25(4):651–9. doi: 10.1093/hmg/ddv502 WOS:000372151000003. 26662796

22. Carbone MA, Yamamoto A, Huang W, Lyman RA, Meadors TB, Yamamoto R, et al. Genetic architecture of natural variation in visual senescence in Drosophila. Proceedings of the National Academy of Sciences of the United States of America. 2016;113(43):E6620–E9. doi: 10.1073/pnas.1613833113 WOS:000386087100012. 27791033

23. Zhou SS, Morozova TV, Hussain YN, Luoma SE, McCoy L, Yamamoto A, et al. The Genetic Basis for Variation in Sensitivity to Lead Toxicity in Drosophila melanogaster. Environmental Health Perspectives. 2016;124(7):1062–70. doi: 10.1289/ehp.1510513 WOS:000380749300029. 26859824

24. Montgomery SL, Vorojeikina D, Huang W, Mackay TFC, Anholt RRH, Rand MD. Genome-Wide Association Analysis of Tolerance to Methylmercury Toxicity in Drosophila Implicates Myogenic and Neuromuscular Developmental Pathways. Plos One. 2014;9(10). doi: 10.1371/journal.pone.0110375 WOS:000343943700033. 25360876

25. Nakka P, Raphael BJ, Ramachandran S. Gene and Network Analysis of Common Variants Reveals Novel Associations in Multiple Complex Diseases. Genetics. 2016;204(2):783–+. doi: 10.1534/genetics.116.188391 WOS:000385871400031. 27489002

26. Babcock DT, Ganetzky B. An Improved Method for Accurate and Rapid Measurement of Flight Performance in Drosophila. Jove-Journal of Visualized Experiments. 2014;(84). doi: 10.3791/51223 WOS:000348604100055. 24561810

27. Carbon S, Douglass E, Dunn N, Good B, Harris NL, Lewis SE, et al. The Gene Ontology Resource: 20 years and still GOing strong. Nucleic Acids Research. 2019;47(D1):D330–D8. doi: 10.1093/nar/gky1055 WOS:000462587400049. 30395331

28. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene Ontology: tool for the unification of biology. Nature Genetics. 2000;25(1):25–9. doi: 10.1038/75556 WOS:000086884000011. 10802651

29. Mackay TF, Huang W. Charting the genotype–phenotype map: lessons from the Drosophila melanogaster Genetic Reference Panel. Wiley Interdisciplinary Reviews: Developmental Biology. 2018;7(1). doi: 10.1002/wdev.289 28834395

30. Pitchers W, Nye J, Marquez EJ, Kowalski A, Dworkin I, Houle D. A Multivariate Genome-Wide Association Study of Wing Shape in Drosophila melanogaster. Genetics. 2019;211(4):1429–47. doi: 10.1534/genetics.118.301342 WOS:000463935700021. 30792267

31. Paul L, Wang SH, Manivannan SN, Bonanno L, Lewis S, Austin CL, et al. Dpp-induced Egfr signaling triggers postembryonic wing development in Drosophila. Proceedings of the National Academy of Sciences of the United States of America. 2013;110(13):5058–63. doi: 10.1073/pnas.1217538110 WOS:000318031900048. 23479629

32. Crossman SH, Streichan SJ, Vincent JP. EGFR signaling coordinates patterning with cell survival during Drosophila epidermal development. Plos Biology. 2018;16(10). doi: 10.1371/journal.pbio.3000027 WOS:000449322300040. 30379844

33. Sibilia M, Kroismayr R, Lichtenberger BM, Natarajan A, Hecking M, Holcmann M. The epidermal growth factor receptor: from development to tumorigenesis. Differentiation. 2007;75(9):770–87. doi: 10.1111/j.1432-0436.2007.00238.x WOS:000250816400002. 17999740

34. Li XY, MacArthur S, Bourgon R, Nix D, Pollard DA, Iyer VN, et al. Transcription factors bind thousands of active and inactive regions in the Drosophila blastoderm. Plos Biology. 2008;6(2):365–88. doi: 10.1371/journal.pbio.0060027 WOS:000254928400023. 18271625

35. MacArthur S, Li XY, Li JY, Brown JB, Chu HC, Zeng L, et al. Developmental roles of 21 Drosophila transcription factors are determined by quantitative differences in binding to an overlapping set of thousands of genomic regions. Genome Biology. 2009;10(7). doi: 10.1186/gb-2009-10-7-r80 WOS:000269309600015. 19627575

36. Moses AM, Pollard DA, Nix DA, Iyer VN, Li XY, Biggin MD, et al. Large-scale turnover of functional transcription factor binding sites in Drosophila. Plos Computational Biology. 2006;2(10):1219–31. doi: 10.1371/journal.pcbi.0020130 WOS:000241921500006. 17040121

37. Thomas S, Li XY, Sabo PJ, Sandstrom R, Thurman RE, Canfield TK, et al. Dynamic reprogramming of chromatin accessibility during Drosophila embryo development. Genome Biology. 2011;12(5). doi: 10.1186/gb-2011-12-5-r43 WOS:000295732700007. 21569360

38. dos Santos G, Schroeder AJ, Goodman JL, Strelets VB, Crosby MA, Thurmond J, et al. FlyBase: introduction of the Drosophila melanogaster Release 6 reference genome assembly and large-scale migration of genome annotations. Nucleic Acids Research. 2015;43(D1):D690–D7. doi: 10.1093/nar/gku1099 WOS:000350210400101. 25398896

39. Grumbling G, Strelets V, FlyBase C. FlyBase: anatomical data, images and queries. Nucleic Acids Research. 2006;34:D484–D8. doi: 10.1093/nar/gkj068 WOS:000239307700106. 16381917

40. Kofler R, Schlötterer C. Gowinda: unbiased analysis of gene set enrichment for genome-wide association studies. Bioinformatics. 2012;28(15):2084–5. doi: 10.1093/bioinformatics/bts315 22635606

41. Metaxakis A, Oehler S, Klinakis A, Savakis C. Minos as a genetic and genomic tool in Drosophila melanogaster. Genetics. 2005;171(2):571–81. doi: 10.1534/genetics.105.041848 WOS:000233194500014. 15972463

42. Bellen HJ, Levis RW, He YC, Carlson JW, Evans-Holm M, Bae E, et al. The Drosophila Gene Disruption Project: Progress Using Transposons With Distinctive Site Specificities. Genetics. 2011;188(3):731–U341. doi: 10.1534/genetics.111.126995 WOS:000292538900022. 21515576

43. Mummery-Widmer JL, Yamazaki M, Stoeger T, Novatchkova M, Bhalerao S, Chen D, et al. Genome-wide analysis of Notch signalling in Drosophila by transgenic RNAi. Nature. 2009;458(7241):987–U59. doi: 10.1038/nature07936 WOS:000265412900033. 19363474

44. Neumuller RA, Richter C, Fischer A, Novatchkova M, Neumuller KG, Knoblich JA. Genome-Wide Analysis of Self-Renewal in Drosophila Neural Stem Cells by Transgenic RNAi. Cell Stem Cell. 2011;8(5):580–93. doi: 10.1016/j.stem.2011.02.022 WOS:000290927600017. 21549331

45. Firth LC, Baker NE. Spitz from the retina regulates genes transcribed in the second mitotic wave, peripodial epithelium, glia and plasmatocytes of the Drosophila eye imaginal disc. Developmental Biology. 2007;307(2):521–38. doi: 10.1016/j.ydbio.2007.04.037 WOS:000248019100027. 17553483

46. Hummel T, Zipursky L. Afferent induction of olfactory glomeruli requires N-cadherin. Neuron. 2004;42(1):77–88. doi: 10.1016/s0896-6273(04)00158-8 WOS:000221458400009. 15066266

47. Hummel T, Vasconcelos ML, Clemens JC, Fishilevich Y, Vosshall LB, Zipursky SL. Axonal targeting of olfactory receptor neurons in Drosophila is controlled by Dscam. Neuron. 2003;37(2):221–31. doi: 10.1016/s0896-6273(02)01183-2 WOS:000181054900008. 12546818

48. Soba P, Zhu S, Emoto K, Younger S, Yang SJ, Yu HH, et al. Drosophila sensory neurons require Dscam for dendritic self-avoidance and proper dendritic field organization. Neuron. 2007;54(3):403–16. doi: 10.1016/j.neuron.2007.03.029 WOS:000246855700011. 17481394

49. Tadros W, Xu SW, Akin O, Yi CH, Shin GJE, Millard SS, et al. Dscam Proteins Direct Dendritic Targeting through Adhesion. Neuron. 2016;89(3):480–93. doi: 10.1016/j.neuron.2015.12.026 WOS:000373564900009. 26844831

50. Luo K. Signaling cross talk between TGF-β/Smad and other signaling pathways. Cold Spring Harbor perspectives in biology. 2017;9(1):a022137. doi: 10.1101/cshperspect.a022137 27836834

51. Krupp JJ, Yaich LE, Wessells RJ, Bodmer R. Identification of genetic loci that interact with cut during Drosophila wing-margin development. Genetics. 2005;170(4):1775–95. doi: 10.1534/genetics.105.043125 WOS:000232033300028. 15956666

52. Fernandes I, Schock F. The nebulin repeat protein Lasp regulates I-band architecture and filament spacing in myofibrils. Journal of Cell Biology. 2014;206(4):559–72. doi: 10.1083/jcb.201401094 WOS:000340739500009. 25113030

53. Spletter ML, Barz C, Yeroslaviz A, Schonbauer C, Ferreira IRS, Sarov M, et al. The RNA-binding protein Arrest (Bruno) regulates alternative splicing to enable myofibril maturation in Drosophila flight muscle. Embo Reports. 2015;16(2):178–91. doi: 10.15252/embr.201439791 WOS:000350695600009. 25532219

54. Kowalewski-Nimmerfall E, Schahs P, Maresch D, Rendic D, Kramer H, Mach L. Drosophila melanogaster cellular repressor of E1A-stimulated genes is a lysosomal protein essential for fly development. Biochimica Et Biophysica Acta-Molecular Cell Research. 2014;1843(12):2900–12. doi: 10.1016/j.bbamcr.2014.08.012 WOS:000343785700010. 25173815

55. Reyna MA, Leiserson MDM, Raphael BJ. Hierarchical HotNet: identifying hierarchies of altered subnetworks. Bioinformatics. 2018;34(17):972–80. doi: 10.1093/bioinformatics/bty613 WOS:000444317200045. 30423088

56. Huang W, Richards S, Carbone MA, Zhu DH, Anholt RRH, Ayroles JF, et al. Epistasis dominates the genetic architecture of Drosophila quantitative traits. Proceedings of the National Academy of Sciences of the United States of America. 2012;109(39):15553–9. doi: 10.1073/pnas.1213423109 WOS:000309604500014. 22949659

57. Adams CM, Anderson MG, Motto DG, Price MP, Johnson WA, Welsh MJ. Ripped pocket and pickpocket, novel Drosophila DEG/ENaC subunits expressed in early development and in mechanosensory neurons. Journal of Cell Biology. 1998;140(1):143–52. doi: 10.1083/jcb.140.1.143 WOS:000071500600014. 9425162

58. Lu BK, LaMora A, Sun YS, Welsh MJ, Ben-Shahar Y. ppk23-Dependent Chemosensory Functions Contribute to Courtship Behavior in Drosophila melanogaster. Plos Genetics. 2012;8(3). doi: 10.1371/journal.pgen.1002587 WOS:000302254800062. 22438833

59. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9(1):559. doi: 10.1186/1471-2105-9-559 19114008

60. Huang W, Carbone MA, Magwire MM, Peiffer JA, Lyman RF, Stone EA, et al. Genetic basis of transcriptome diversity in Drosophila melanogaster. Proceedings of the National Academy of Sciences. 2015;112(44):E6010–E9. doi: 10.1073/pnas.1519159112 26483487

61. Crawford L, Zeng P, Mukherjee S, Zhou X. Detecting epistasis with the marginal epistasis test in genetic mapping studies of quantitative traits. Plos Genetics. 2017;13(7). doi: 10.1371/journal.pgen.1006869 WOS:000406615300011. 28746338

62. Fuerst PG, Burgess RW. Adhesion molecules in establishing retinal circuitry. Current Opinion in Neurobiology. 2009;19(4):389–94. doi: 10.1016/j.conb.2009.07.013 WOS:000270107600007. 19660931

63. Emoto K, He Y, Ye B, Grueber WB, Adler PN, Jan LY, et al. Control of dendritic branching and tiling by the tricornered-kinase/furry signaling pathway in Drosophila sensory neurons. Cell. 2004;119(2):245–56. doi: 10.1016/j.cell.2004.09.036 WOS:000224577200012. 15479641

64. Matsubara D, Horiuchi SY, Shimono K, Usui T, Uemura T. The seven-pass transmembrane cadherin Flamingo controls dendritic self-avoidance via its binding to a LIM domain protein, Espinas, in Drosophila sensory neurons. Genes & Development. 2011;25(18):1982–96. doi: 10.1101/gad.16531611 WOS:000295082400009. 21937715

65. Zhan XL, Clemens JC, Neves G, Hattori D, Flanagan JJ, Hummel T, et al. Analysis of Dscam diversity in regulating axon guidance in Drosophila mushroom bodies. Neuron. 2004;43(5):673–86. doi: 10.1016/j.neuron.2004.07.020 WOS:000223691900013. 15339649

66. Neves G, Zucker J, Daly M, Chess A. Stochastic yet biased expression of multiple Dscam splice variants by individual cells. Nature Genetics. 2004;36(3):240–6. doi: 10.1038/ng1299 WOS:000189250400012. 14758360

67. Thistle R, Cameron P, Ghorayshi A, Dennison L, Scott K. Contact chemoreceptors mediate male-male repulsion and male-female attraction during Drosophila courtship. Cell. 2012;149(5):1140–51. doi: 10.1016/j.cell.2012.03.045 22632976

68. Guo Y, Wang Y, Wang Q, Wang Z. The role of PPK26 in Drosophila larval mechanical nociception. Cell reports. 2014;9(4):1183–90. doi: 10.1016/j.celrep.2014.10.020 25457610

69. Gorczyca DA, Younger S, Meltzer S, Kim SE, Cheng L, Song W, et al. Identification of Ppk26, a DEG/ENaC channel functioning with Ppk1 in a mutually dependent manner to guide locomotion behavior in Drosophila. Cell reports. 2014;9(4):1446–58. doi: 10.1016/j.celrep.2014.10.034 25456135

70. Elkins T, Ganetzky B, Wu CF. A DROSOPHILA MUTATION THAT ELIMINATES A CALCIUM-DEPENDENT POTASSIUM CURRENT. Proceedings of the National Academy of Sciences of the United States of America. 1986;83(21):8415–9. doi: 10.1073/pnas.83.21.8415 WOS:A1986E643100082. 2430288

71. Homyk T, Grigliatti TA. BEHAVIORAL MUTANTS OF DROSOPHILA-MELANOGASTER .4. ANALYSIS OF DEVELOPMENTALLY TEMPERATURE-SENSITIVE MUTATIONS AFFECTING FLIGHT. Developmental Genetics. 1983;4(2):77–97. doi: 10.1002/dvg.1020040204 WOS:A1983SA35500003.

72. Homyk T, Szidonya J, Suzuki DT. BEHAVIORAL MUTANTS OF DROSOPHILA-MELANOGASTER .3. ISOLATION AND MAPPING OF MUTATIONS BY DIRECT VISUAL OBSERVATIONS OF BEHAVIORAL PHENOTYPES. Molecular & General Genetics. 1980;177(4):553–65. WOS:A1980JM30700002.

73. Iyengar A, Wu CF. Flight and Seizure Motor Patterns in Drosophila Mutants: Simultaneous Acoustic and Electrophysiological Recordings of Wing Beats and Flight Muscle Activity. Journal of Neurogenetics. 2014;28(3–4):316–28. doi: 10.3109/01677063.2014.957827 WOS:000344555700015. 25159538

74. Wilson DM. The central nervous control of flight in a locust. Journal of Experimental Biology. 1961;38(2):471–90.

75. Ainsley JA, Kim MJ, Wegman LJ, Pettus JM, Johnson WA. Sensory mechanisms controlling the timing of larval developmental and behavioral transitions require the Drosophila DEG/ENaC subunit, Pickpocket1. Developmental Biology. 2008;322(1):46–55. doi: 10.1016/j.ydbio.2008.07.003 WOS:000259790000005. 18674528

76. Ainsley JA, Pettus JM, Bosenko D, Gerstein CE, Zinkevich N, Anderson MG, et al. Enhanced locomotion caused by loss of the Drosophila DEG/ENaC protein pickpocket1. Current Biology. 2003;13(17):1557–63. doi: 10.1016/s0960-9822(03)00596-7 WOS:000185171300028. 12956960

77. Orr BO, Gorczyca D, Younger MA, Jan LY, Jan YN, Davis GW. Composition and Control of a Deg/ENaC Channel during Presynaptic Homeostatic Plasticity. Cell Reports. 2017;20(8):1855–66. doi: 10.1016/j.celrep.2017.07.074 WOS:000408154300011. 28834749

78. Younger MA, Muller M, Tong A, Pym EC, Davis GW. A Presynaptic ENaC Channel Drives Homeostatic Plasticity. Neuron. 2013;79(6):1183–96. doi: 10.1016/j.neuron.2013.06.048 WOS:000330329600015. 23973209

79. Kimura KI, Ote M, Tazawa T, Yamamoto D. Fruitless specifies sexually dimorphic neural circuitry in the Drosophila brain. Nature. 2005;438(7065):229–33. doi: 10.1038/nature04229 WOS:000233133500050. 16281036

80. Northcutt AJ, Schulz DJ. Molecular mechanisms of homeostatic plasticity in central pattern generator networks. Developmental Neurobiology. 2020;80(1–2):58–69. doi: 10.1002/dneu.22727 WOS:000502572300001. 31778295

81. Ben-Shahar Y. Sensory Functions for Degenerin/Epithelial Sodium Channels (DEG/ENaC). In: Friedmann T, Dunlap JC, Goodwin SF, editors. Advances in Genetics, Vol 76. Advances in Genetics. 762011. p. 1–26. doi: 10.1016/B978-0-12-386481-9.00001-8 22099690

82. Gendron CM, Kuo TH, Harvanek ZM, Chung BY, Yew JY, Dierick HA, et al. Drosophila Life Span and Physiology Are Modulated by Sexual Perception and Reward. Science. 2014;343(6170):544–8. doi: 10.1126/science.1243339 WOS:000330343700048. 24292624

83. Yu JY, Kanai MI, Demir E, Jefferis G, Dickson BJ. Cellular Organization of the Neural Circuit that Drives Drosophila Courtship Behavior. Current Biology. 2010;20(18):1602–14. doi: 10.1016/j.cub.2010.08.025 WOS:000282385600020. 20832315

84. Shirangi TR, Wong AM, Truman JW, Stern DL. Doublesex Regulates the Connectivity of a Neural Circuit Controlling Drosophila Male Courtship Song. Developmental Cell. 2016;37(6):533–44. doi: 10.1016/j.devcel.2016.05.012 WOS:000378204200008. 27326931

85. Rezaval C, Pavlou HJ, Dornan AJ, Chan YB, Kravitz EA, Goodwin SF. Neural Circuitry Underlying Drosophila Female Postmating Behavioral Responses. Current Biology. 2012;22(13):1155–65. doi: 10.1016/j.cub.2012.04.062 WOS:000306379600018. 22658598

86. Mardahl-Dumesnil M, Fowler VM. Thin filaments elongate from their pointed ends during myofibril assembly in Drosophila indirect flight muscle. Journal of Cell Biology. 2001;155(6):1043–53. doi: 10.1083/jcb.200108026 WOS:000172730200017. 11739412

87. Hevia CF, Lopez-Varea A, Esteban N, de Celis JF. A Search for Genes Mediating the Growth-Promoting Function of TGF beta in the Drosophila melanogaster Wing Disc. Genetics. 2017;206(1):231–49. doi: 10.1534/genetics.116.197228 WOS:000401127800016. 28315837

88. Cruz C, Glavic A, Casado M, de Celis JF. A Gain-of-Function Screen Identifying Genes Required for Growth and Pattern Formation of the Drosophila melanogaster Wing. Genetics. 2009;183(3):1005–26. doi: 10.1534/genetics.109.107748 WOS:000272295800021. 19737745

89. Yu K, Sturtevant MA, Biehs B, Francois V, Padgett RW, Blackman RK, et al. The Drosophila decapentaplegic and short gastrulation genes function antagonistically during adult wing vein development. Development. 1996;122(12):4033–44. WOS:A1996WC55400034. 9012523

90. Bangi E, Wharton K. Dual function of the Drosophila Alk1/Alk2 ortholog Saxophone shapes the Bmp activity gradient in the wing imaginal disc. Development. 2006;133(17):3295–303. doi: 10.1242/dev.02513 16887821

91. Bangi E, Wharton K. Dpp and Gbb exhibit different effective ranges in the establishment of the BMP activity gradient critical for Drosophila wing patterning. Developmental biology. 2006;295(1):178–93. doi: 10.1016/j.ydbio.2006.03.021 16643887

92. Ball RW, Warren-Paquin M, Tsurudome K, Liao EH, Elazzouzi F, Cavanagh C, et al. Retrograde BMP Signaling Controls Synaptic Growth at the NMJ by Regulating Trio Expression in Motor Neurons. Neuron. 2010;66(4):536–49. doi: 10.1016/j.neuron.2010.04.011 WOS:000278771100009. 20510858

93. Fischer S, Bayersdorfer F, Harant E, Reng R, Arndt S, Bosserhoff AK, et al. fussel (fuss)—A Negative Regulator of BMP Signaling in Drosophila melanogaster. Plos One. 2012;7(8). doi: 10.1371/journal.pone.0042349 WOS:000307437900031. 22879948

94. Takaesu NT, Hyman-Walsh C, Ye Y, Wisotzkey RG, Stinchfield MJ, O’Connor MB, et al. dSno facilitates Baboon signaling in the Drosophila brain by switching the affinity of Medea away from Mad and toward dSmad2. Genetics. 2006;174(3):1299–313. doi: 10.1534/genetics.106.064956 16951053

95. O’Connor MB, Umulis D, Othmer HG, Blair SS. Shaping BMP morphogen gradients in the Drosophila embryo and pupal wing. Development. 2006;133(2):183–93. WOS:000235345700001. doi: 10.1242/dev.02214 16368928

96. Wharton KA, Serpe M. Fine-tuned shuttles for bone morphogenetic proteins. Current Opinion in Genetics & Development. 2013;23(4):374–84. doi: 10.1016/j.gde.2013.04.012 WOS:000324362900003. 23735641

97. Wittkopp PJ, Haerum BK, Clark AG. Evolutionary changes in cis and trans gene regulation. Nature. 2004;430(6995):85–8. doi: 10.1038/nature02698 WOS:000222356800048. 15229602

98. Hacker U, Grossniklaus U, Gehring WJ, Jackle H. DEVELOPMENTALLY REGULATED DROSOPHILA GENE FAMILY ENCODING THE FORK HEAD DOMAIN. Proceedings of the National Academy of Sciences of the United States of America. 1992;89(18):8754–8. doi: 10.1073/pnas.89.18.8754 WOS:A1992JN50300071. 1356269

99. Dickinson MH, Lehmann F-O, Sane SP. Wing rotation and the aerodynamic basis of insect flight. Science. 1999;284(5422):1954–60. doi: 10.1126/science.284.5422.1954 10373107

100. Bellen HJ, Levis RW, He Y, Carlson JW, Evans-Holm M, Bae E, et al. The drosophila gene disruption project: progress using transposons with distinctive site-specificities. Genetics. 2011:genetics. 111.126995. doi: 10.1534/genetics.111.126995 21515576

101. Elgin S, Miller D. Special techniques for tissue isolation and injection. II. Mass rearing of flies and mass production and harvesting of embryos. Genetics and biology of Drosophila. 1978.

102. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature Methods. 2012;9(7):676–82. doi: 10.1038/nmeth.2019 WOS:000305942200021. 22743772

103. Das J, Yu HY. HINT: High-quality protein interactomes and their applications in understanding human disease. Bmc Systems Biology. 2012;6. doi: 10.1186/1752-0509-6-92 WOS:000310442400001. 22846459

104. Yu JK, Pacifico S, Liu GZ, Finley RL. DroID: the Drosophila Interactions Database, a comprehensive resource for annotated gene and protein interactions. Bmc Genomics. 2008;9. doi: 10.1186/1471-2164-9-9 WOS:000260678600001. 18186939

105. Murali T, Pacifico S, Yu JK, Guest S, Roberts GG, Finley RL. DroID 2011: a comprehensive, integrated resource for protein, transcription factor, RNA and gene interactions for Drosophila. Nucleic Acids Research. 2011;39:D736–D43. doi: 10.1093/nar/gkq1092 WOS:000285831700117. 21036869

106. Hu YH, Flockhart I, Vinayagam A, Bergwitz C, Berger B, Perrimon N, et al. An integrative approach to ortholog prediction for disease-focused and other functional studies. Bmc Bioinformatics. 2011;12(1):357. doi: 10.1186/1471-2105-12-357 WOS:000295225400001. 21880147

107. Eden E, Navon R, Steinfeld I, Lipson D, Yakhini Z. GOrilla: a tool for discovery and visualization of enriched GO terms in ranked gene lists. Bmc Bioinformatics. 2009;10. doi: 10.1186/1471-2105-10-10 WOS:000264007400001. 19133123

108. Eden E, Lipson D, Yogev S, Yakhini Z. Discovering motifs in ranked lists of DNA sequences. Plos Computational Biology. 2007;3(3):508–22. doi: 10.1371/journal.pcbi.0030039 WOS:000246191000017. 17381235

109. Browning SR, Browning BL. Rapid and accurate haplotype phasing and missing-data inference for whole-genome association studies by use of localized haplotype clustering. American Journal of Human Genetics. 2007;81(5):1084–97. doi: 10.1086/521987 WOS:000250480900018. 17924348

110. Browning BL, Browning SR. Genotype Imputation with Millions of Reference Samples. American Journal of Human Genetics. 2016;98(1):116–26. doi: 10.1016/j.ajhg.2015.11.020 WOS:000368050800009. 26748515

111. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinformatics. 2011;27(15):2156–8. doi: 10.1093/bioinformatics/btr330 WOS:000292778700023. 21653522

112. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. American Journal of Human Genetics. 2007;81(3):559–75. doi: 10.1086/519795 WOS:000249128200012. 17701901

113. Frise E, Hammonds AS, Celniker SE. Systematic image-driven analysis of the spatial Drosophila embryonic expression landscape. Molecular Systems Biology. 2010;6. doi: 10.1038/msb.2009.102 WOS:000274193500002. 20087342

114. Chintapalli VR, Wang J, Dow JAT. Using FlyAtlas to identify better Drosophila melanogaster models of human disease. Nature Genetics. 2007;39(6):715–20. doi: 10.1038/ng2049 WOS:000246859100012. 17534367

115. Consortium TAoGR. Alliance of Genome Resources Portal: unified model organism research platform. Nucleic acids research. 2020;48(D1):D650–D8. doi: 10.1093/nar/gkz813 31552413

116. Gaudet P, Livstone MS, Lewis SE, Thomas PD. Phylogenetic-based propagation of functional annotations within the Gene Ontology consortium. Briefings in Bioinformatics. 2011;12(5):449–62. doi: 10.1093/bib/bbr042 WOS:000295171700010. 21873635

117. Karr TL. Fruit flies and the sperm proteome. Human Molecular Genetics. 2007;16:R124–R33. doi: 10.1093/hmg/ddm252 WOS:000251165500002. 17911156

118. Casas-Vila N, Bluhm A, Sayols S, Dinges N, Dejung M, Altenhein T, et al. The developmental proteome of Drosophila melanogaster. Genome Research. 2017;27(7):1273–85. doi: 10.1101/gr.213694.116 WOS:000404735500016. 28381612

119. Brown JB, Boley N, Eisman R, May GE, Stoiber MH, Duff MO, et al. Diversity and dynamics of the Drosophila transcriptome. Nature. 2014;512(7515):393–9. doi: 10.1038/nature12962 WOS:000340840600025. 24670639


Článek vyšel v časopise

PLOS Genetics


2021 Číslo 3
Nejčtenější tento týden
Nejčtenější v tomto čísle
Kurzy

Zvyšte si kvalifikaci online z pohodlí domova

Důležitost adherence při depresivním onemocnění
nový kurz
Autoři: MUDr. Eliška Bartečková, Ph.D.

Koncepce osteologické péče pro gynekology a praktické lékaře
Autoři: MUDr. František Šenk

Sekvenční léčba schizofrenie
Autoři: MUDr. Jana Hořínková, Ph.D.

Hypertenze a hypercholesterolémie – synergický efekt léčby
Autoři: prof. MUDr. Hana Rosolová, DrSc.

Multidisciplinární zkušenosti u pacientů s diabetem
Autoři: Prof. MUDr. Martin Haluzík, DrSc., prof. MUDr. Vojtěch Melenovský, CSc., prof. MUDr. Vladimír Tesař, DrSc.

Všechny kurzy
Přihlášení
Zapomenuté heslo

Zadejte e-mailovou adresu, se kterou jste vytvářel(a) účet, budou Vám na ni zaslány informace k nastavení nového hesla.

Přihlášení

Nemáte účet?  Registrujte se

#ADS_BOTTOM_SCRIPTS#