Computational protocol: Candidate genes for cooperation and aggression in the social wasp Polistes dominula

Similar protocols

Protocol publication

[…] We designed primers to perform quantitative real-time PCR (qPCR) on candidate genes for cooperation and aggression. We obtained the P. metricus sequences for the candidate genes from study A and we used them to retrieve the orthologue sequences in the P. dominula genome (Standage et al. ). For this purpose, we used the P. dominula genome deposited on the virtual genome annotation platform GDB (version PdomGDB r1.2). The reverse complement of the P. dominula genomic sequence was aligned to the P. metricus candidate gene sequence using the GDB: this step enabled us to select a portion for primer design that was overlapping, at least partially, with the region targeted by the microarray probe in study A. In this way, we made sure that we targeted with our qPCR approach the same exon(s) that coded for differentially expressed mRNAs in the microarray analysis. We used the web-tool Primer3plus ( to design forward and reverse primers for the targeted regions (Supp Table 6). We performed a BLAST screening of the designed primers to make sure that they did not target other regions in the P. dominula genome not associated with the genes of interest. [...] To analyse qPCR data, we obtained Cq values for each wasp phenotype per candidate gene and we averaged across the 3 technical replicates: outliers (i.e., individual replicates that had a Cq value of 0.5 higher or lower than the average) were manually removed beforehand. Averaged values were then normalized against the control genes actin and elongation factor 1 (variance = 0.166 and 0.185, respectively). Normalized gene expression values (Supp Table 9) were analysed in SPSS Statistics (IBM, version 21). We performed a Multivariate Analysis of Variance (MANOVA) across all candidate genes to detect the effect of wasp phenotype on global gene expression, and then we performed LSD post hoc tests to detect significant differences between pairs of phenotypes for each gene that we tested. Finally, we also performed a Correlation analysis to identify patterns of co-expression among candidate genes. We corrected for multiple comparisons using Bonferroni (significance threshold = 0.005).Hierarchical clustering analyses and heatmaps were obtained in R using the packages “gplots” and “RColorBrewer”: normalized gene expression data were scaled beforehand. Analyses of possible effects of fat body rank, ovary size, egg development and wasp location on gene expression were performed using General Linear Models (GLM) in SPSS. We built a model for each factor and added the measures of gaster size (length, width, length × width) as covariates in the design (Supp Table 10). We corrected for multiple comparisons using Bonferroni (significance threshold = 0.012).To build the network of global gene expression in the P. dominula head we obtained expression levels from a head RNA-seq experiment described in Standage et al. (). We restricted the starting material to the 6298 genes that were queen-biased in that study, i.e. genes that showed more than onefold higher expression levels in queen vs. worker heads. We performed weighted gene co-expression network analysis (WGCNA) using a similar protocol as in Manfredini et al. () with default settings and soft thresholding power = 30. We used VisANT (Hu et al. ), to visualize the network structure.Gene Ontology (GO) analyses were performed with Blast2GO (version 4.1.9). Functional Annotation Clustering of GO terms resulting from the overlap between study A and study B was performed in DAVID (Huang et al. ) considering only “biological processes” and using the default Drosophila population background. GO terms associated with vitellogenin-correlated genes were visualized using REVIGO (Supek et al. ). We restricted this analysis to “biological processes” only and to GO terms obtained through the InterPRO database. […]

Pipeline specifications

Software tools Primer3, SPSS, gplots, Blast2GO, DAVID, REViGO
Applications Miscellaneous, RNA-seq analysis, qPCR
Organisms Polistes dominula