Multi-genotype analyses of long-ncRNAs in Sugarcane

Identification of long non-coding RNAs in sugarcane

• Predicting putative lncRNAs with CPC2
• Predicting putative lncRNAs with PLncPRO
• Predicting putative lncRNAs with RNAplonc
• Aggregating the consensus of putative ncRNAs predicted by the 3 tools

The following results represent the percentage of sequences classified as lncRNAs from the 3 softwares used in identifying putative lncRNAs in sugarcane (described in detail above). The 16,237,098 input sequences originate from 48 sugarcane transcriptomes that were assembled and analyzed in a previous project, where we conducted inference and annotation of the sugarcane pan-transcriptome.

Families/classes of lncrnas

• Annotation of ncRNAs families

Conserved long non-coding RNAs in the panRNAome

• Clustering sequences by identity
• Inference of the pan-ncRNAome

Sugarcane lncRNAs origin

• Estimating lncRNA origin

Co-expression patterns of lncRNAs

• Contrasting genotypes selection
• Calculating the RNA expression for each sample
• Filtering the RNAs expression matrix of the contrasting genotypes
• Filtering the RNAs expression matrix by function
• Analyzing RNA expression patterns after filtering
• Principal Component Analysis of lncRNAs expression in contrasting genotypes
• Computing gene pair relationships with Pearson correlation
• Analyzing Network Metrics
• Analyzing MCL clusterings (effects of Inflation value)
• Finding conserved modules of co-expression (complete subgraphs - cliques)

• Visualizing co-expressed gene modules in contrasting genotypes

• Gene Ontology enrichment analysis: exploring biological function of modules