Weighted gene co-expression network analysis identifies modules and functionally enriched pathways in the lactation process

Abstract

The exponential growth in knowledge has resulted in a better understanding of the lactation process in a wide variety of animals. However, the underlying genetic mechanisms are not yet clearly known. In order to identify the mechanisms involved in the lactation process, various mehods, including meta-analysis, weighted gene co-express network analysis (WGCNA), hub genes identification, gene ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment at before peak (BP), peak (P), and after peak (AP) stages of the lactation processes have been employed. A total of 104, 85, and 26 differentially expressed genes were identified based on PB vs. P, BP vs. AP, and P vs. AP comparisons, respectively. GO and KEGG pathway enrichment analysis revealed that DEGs were significantly enriched in the "ubiquitin-dependent ERAD" and the "chaperone cofactor-dependent protein refolding" in BP vs. P and P vs. P, respectively. WGCNA identified five significant functional modules related to the lactation process. Moreover, GJA1, AP2A2, and NPAS3 were defined as hub genes in the identified modules, highlighting the importance of their regulatory impacts on the lactation process. The findings of this study provide new insights into the complex regulatory networks of the lactation process at three distinct stages, while suggesting several candidate genes that may be useful for future animal breeding programs. Furthermore, this study supports the notion that in combination with a meta-analysis, the WGCNA represents an opportunity to achieve a higher resolution analysis that can better predict the most important functional genes that might provide a more robust bio-signature for phenotypic traits, thus providing more suitable biomarker candidates for future studies.

Figure 1

Network visualization of enriched pathways (GO/KEGG) in the gene signature was performed by ClueGO analysis. (A) BP vs. P and (B) P vs. AP.

Figure 2

Weighted gene co-expression network analysis (WGCNA) of (A) the hierarchical cluster tree of 13,591 meta-genes between the three species. The branches and color bands represent the assigned module; and (B) co-expression network modules. In the Topological Overlap Matrix (TOM) plot, the light color represents low overlap and the progressively darker red color represents higher overlap between the genes.

Figure 3

The 17 modules identified by the weighted co-expression analysis (WGCNA) along with the number of genes in each module.

Figure 4

The module trait relationship (p value) for identified modules (y-axis) in relation with traits (x-axis). X-axis legend: BP = before peak; P = Peak; AP = after peak.

Figure 5

Gene networks for DEGs involved in the lactation process. (A) BP vs. P; (B) BP vs. AP; and (C) P vs. AP comparisons are shown. The mapping strategy of using low parameter values corresponding to bright colors was used for node coloring. The brightest color is green and the darkest color is red. The default middle color is yellow.

Figure 6

The Venn diagram representing the number of DEGs selected by the meta-analysis and the number of genes selected by the significant modules in the weighted co-expression analysis.

Figure 7

Graphical model of decision tree using Information Gain criterion based on hub genes in three different stages of lactation (Before Peak (BP), Peak (P), and After Peak (AP)).

Figure 8

Flowchart of the performed meta-analysis and WGCNA analysis of the lactation process using the RNA-Seq datasets.