RNA-seq Data Processing by HPC

Quality Check - FastQC

Read Mapping - HISAT

Quantification, Normalization of "transcript-level" - Cufflinks and Cuffnorm

quantifying "Gene-level Expression" - FeatureCounts

Project Overview:

In the realm of genomics research, RNA-seq has emerged as a transformative technology, empowering us to explore "Gene Expression" intricacies with unprecedented depth. This project demonstrates essential skills for tackling challenges and seizing opportunities in RNA-seq analysis. Through curated hands-on experiences, we delve into theory and practice, utilizing an adapted dataset from Shaw et al. (2015) to immerse ourselves in the dynamic realm of RNA-seq analysis. The dataset's "treatment" and "control" samples hold keys to unravel the effects of specific drug treatments, guiding us through RNA-seq intricacies and showcasing its power in decoding molecular responses.

• Step 1: Check FASTQ Quality Control

In the pursuit of reliable and accurate genomics analysis, quality control is foundational. We begin by assessing the quality of the sequencing data through the application of FastQC.

Command:

fastqc -o output_directory input_fastq_file(s)

• Step 2: Build HISAT Index

The foundation for alignment by building the HISAT index, an component of the analysis pipeline.

Command:

hisat2-build reference_genome.fa index_prefix

• Step 3: Execute HISAT Alignment

Accurate read mapping is critical to understanding the intricacies of transcription factor binding. Consistent with the methodology of the study, we employ HISAT for its proficiency in aligning reads to the reference genome.

Command:

hisat2 -x index_prefix -U input_fastq_file(s) -S aligned_reads.sam

• Step 4: Conduct Transcript Assembly with Cufflinks

  Cufflinks is responsible for transcript assembly and quantification.

Command:

cufflinks -o output_directory -p num_threads aligned_reads.sam

• Step 5: Normalize Transcript Abundances with Cuffnorm

Normalizing the "transcript-level" data across samples to make them suitable for meaningful comparisons.

Command:

cuffnorm -o output_directory --library-norm-method quartile normalized_data_table.txt

• Step 6: Quantify Gene-level Expression with FeatureCounts

FeatureCounts is a critical tool for quantifying "gene-level expression" from RNA-seq data. It aids in differential expression analysis, functional annotation, and other downstream analyses by providing accurate read counts for genes and other genomic features, enabling researchers to gain insights into the underlying biology of their experimental systems.

Command:

featureCounts -T num_threads -a annotation.gtf -o gene_counts.txt bam_file(s)

Summary:

In this immersive exploration, we navigate the intricate landscape of RNA-seq analysis, unraveling the molecular intricacies that underlie gene expression patterns. Through careful data processing and interpretation, we unlock the potential to uncover transformative insights within genomic data.

References:

Shaw, P., Chaotheing, S., Kaewprommal, P., Piriyapongsa, J., Wongsombat, C., Suwannakitti, N., Koonyosying, P., Uthaipibull, C., Yuthavong, Y., & Kamchonwongpaisan, S. (2015). Plasmodium parasites mount an arrest response to dihydroartemisinin, as revealed by whole transcriptome shotgun sequencing (RNA-seq) and microarray study. BMC Genomics, 16(1). https://doi.org/10.1186/s12864-015-2040-0

Blog:

https://ssidmarine.wordpress.com/2023/08/12/rna-seq-data-processing-by-hpc/

Access data:

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE62136

Some output data available in the data folder