miRNA_seq meta-analysis

This repository contains the miRNA Analysis Pipeline, developed to analyze publicly available NGS datasets (miRNA-seq) for various cancer types. The workflow is implemented using automated scripts and R packages for preprocessing, differential expression analysis, meta-analysis, and functional analysis.

Key Features

1. Preprocessing and Alignment

   • The script mirna.sh performs the following tasks:
     • Quality check of raw reads using FastQC.
     • Trimming of adapters and low-quality bases using fastp or similar tools.
     • Alignment of reads to the genome and read counts using miRDeep2.

2. Differential Expression Analysis

   • Differential expression analysis is conducted on case vs. control datasets from six bioprojects.
   • The results include:
     • Log Fold Change (LFC) values.
     • Standard error estimates for each miRNA.

3. Meta-Analysis

   • Meta-analysis of miRNAs is conducted using the metafor package in R.
   • Filtering criteria:
     • miRNAs must be detected in multiple studies.
     • Log Fold Change (LFC) and standard error ratio (SE) are used to refine results.
   • Outputs include forest plots for selected miRNAs.

4. Functional Analysis

   • Target genes of significant miRNAs are identified using tools such as miRDB.
   • Gene ontology (GO) term enrichment is performed using clusterProfiler.
   • GO terms are visualized with ggplot2.

Tools and Packages Used

• Shell Scripting: Automated preprocessing and alignment steps.
• R Packages:
  • DESeq2 for differential expression analysis.
  • metafor for meta-analysis.
  • clusterProfiler for functional enrichment analysis.
  • ggplot2 for data visualization.

Workflow Overview

1. Preprocessing: Quality check, trimming, and alignment.
2. Differential Expression Analysis: Case vs. control comparisons for individual datasets.
3. Meta-Analysis: Integration of results across multiple studies.
4. Functional Analysis: GO term enrichment and visualization.