All the cleaned reads were mapped on the assembled C. socialis transcriptome using the Bowtie2 aligner (default settings, [64 (link)]). Reads count and FPKM calculation per tag for each replicate was performed using the eXpress software ([65 ]). DEGs calling was performed using two tools implementing two different statistical approaches: DESeq2 ([66 ]) and edgeR ([67 (link)]). The mean of the log2 FC values (log2(FC)) obtained with the two tools was calculated for each transcript. The thresholds for the DEGs calling were FDR ≤ 0.05, P-adjusted ≤ 0.05, and log2(FC) ≥ 1.5|. The union of the DEGs detected by both programs was retained.
A Gene Ontology enrichment analysis of the detected DEGs was performed with Ontologizer software ([68 ]). The threshold used to identify significantly enriched functional terms was P ≤ 0.05. Genes known to be related to different metabolic pathways were manually searched within the transcriptome considering their SwissProt annotation.
A comparative analysis was performed between the transcriptomes of C. socialis and T. pseudonana, which was also studied in nitrogen-limited experimental conditions ([23 (link)]) in non-axenic conditions. The prediction of orthologs was carried out by COMPARO, an in-house software written in python programming language ([69 (link)]).
A Gene Ontology enrichment analysis of the detected DEGs was performed with Ontologizer software ([68 ]). The threshold used to identify significantly enriched functional terms was P ≤ 0.05. Genes known to be related to different metabolic pathways were manually searched within the transcriptome considering their SwissProt annotation.
A comparative analysis was performed between the transcriptomes of C. socialis and T. pseudonana, which was also studied in nitrogen-limited experimental conditions ([23 (link)]) in non-axenic conditions. The prediction of orthologs was carried out by COMPARO, an in-house software written in python programming language ([69 (link)]).
Free full text: Click here
