miRNA mRNA interaction studies

Question

[anjuraas]

hi I want to study miRNA-mRNA interaction in gastric cancer. I downloaded the datasets from SRA and did differential expression using DESEQ2.My pipeline was Bowtie>featurecounts>deseq2. I used NCBI human reference genome file and annotation file. I worked till this using galaxy.Here after I don't know how to proceed. Whether I should go for functional enrichment or functional annotation?what is the next step?Please help

[shawn.w.foley]

I think we need a bit more information about your project and your goals. 1) What are the samples? Drug treated? Mutant vs Wildtype? miRNA upregulation? 2) Is this some sort of miRNA-seq or mRNA-seq? Are you up/downregulating an miRNA then seeing which genes are differentially expressed? 3) Is the goal of the study to see which mRNAs a specific miRNA directly targets? Are you interested in secondary effects?

Sure, you can look at functional enrichment (if you're working on human data then I'd recommend MSigDB and GSEA/GO term analyses), but what do you want out of this data? Do you have a hypothesis? Are you interested in specific genes or pathways?

[anjuraas]

I am doing differential expression analysis using miRNA seq in gastric cancer.i am using 3 sets,control,adjacent,tumor.i have got the deseq2 results.using the miRNAs I want to study angiogenesis is gastric cancer.

[k.kathirvel93]

I am telling this based on the understanding of your Aim. Though you need to perform miRNA-mRNA interaction, next you have to do a target gene prediction of your miRNAs and correlate them with your DE genes from mRNA analysis results. Then go for an Gene enrichment and pathway analysis.

[anjuraas]

Thank you for your reply.I have one more doubt,after doing deseq2 I got the results.but gene ids are given in the table.how can I know the gene names from that?should I search that manually in ncbi(there is nearly 1500 gene ids),or is there any tools for that?

[k.kathirvel93]

Use -g gene_name instead of gene ID while running featureCounts then go for DESeq2.

[anjuraas]

ok.let me try this.thank you

[anjuraas]

I used NCBI humane reference genome.in that gene identifier is ID.

[k.kathirvel93]

Your .gtf annotation file, must have the gene names. Please ensure that.