💻Computational Biology Unit 5 – RNA-Seq Analysis in Transcriptomics

What's RNA-Seq?

• RNA sequencing (RNA-Seq) is a high-throughput sequencing technology used to measure the presence and quantity of RNA in a biological sample at a given moment
• Provides a snapshot of the transcriptome, which is the complete set of RNA transcripts produced by the genome under specific circumstances or in a specific cell
• Enables researchers to analyze the continuously changing cellular transcriptome
• RNA-Seq data offers insight into alternative splicing events, post-transcriptional modifications, gene fusion, mutations/SNPs, and changes in gene expression
• Can be used to determine differential expression of genes under different conditions (treated vs. control) or in different cell types
• Allows for the identification of novel transcripts and isoforms
• Offers a broader dynamic range compared to microarray technology, making it more sensitive for detecting low-abundance transcripts and small changes in expression levels

Key Concepts in Transcriptomics

• Transcriptomics is the study of the transcriptome—the complete set of RNA transcripts that are produced by the genome, under specific circumstances or in a specific cell
• Central dogma of molecular biology: DNA is transcribed into RNA, which is then translated into proteins
• RNA types include messenger RNA (mRNA), ribosomal RNA (rRNA), transfer RNA (tRNA), and various non-coding RNAs (ncRNAs) such as microRNAs (miRNAs) and long non-coding RNAs (lncRNAs)
  • mRNA carries genetic information from DNA to ribosomes for protein synthesis
  • rRNA is a component of ribosomes and plays a role in protein synthesis
  • tRNA transfers specific amino acids to growing polypeptide chains during protein synthesis
  • ncRNAs are involved in various cellular processes, including gene regulation and epigenetic modifications
• Gene expression is the process by which information from a gene is used in the synthesis of a functional gene product, often proteins
• Alternative splicing allows a single gene to produce multiple mRNA isoforms, leading to the production of different protein variants
• Gene regulation controls the timing, location, and amount of gene expression, ensuring that the right genes are expressed at the right time and place

RNA-Seq Workflow Overview

• RNA-Seq workflow typically involves several steps: experimental design, sample preparation, sequencing, data quality control, read alignment, quantification, differential expression analysis, and functional interpretation
• Experimental design considerations include the number of biological replicates, sequencing depth, library preparation methods, and sequencing platform
• Sample preparation involves RNA extraction, quality assessment, and library construction
  • RNA is fragmented, reverse transcribed into cDNA, and adapters are ligated to the ends of the cDNA fragments
• Sequencing is performed using high-throughput sequencing platforms (Illumina, PacBio, Oxford Nanopore)
• Data quality control assesses the quality of raw sequencing reads and filters out low-quality reads and adapters
• Read alignment maps the sequencing reads to a reference genome or transcriptome
• Quantification estimates the expression levels of genes or transcripts based on the aligned reads
• Differential expression analysis identifies genes that are significantly up- or down-regulated between different conditions or groups
• Functional interpretation and pathway analysis help researchers understand the biological significance of the differentially expressed genes

Sample Prep and Sequencing Basics

• RNA extraction is the first step in sample preparation, which involves isolating total RNA from biological samples
  • Methods include organic extraction (TRIzol), silica-based columns, and magnetic bead-based kits
• RNA quality assessment is crucial for successful RNA-Seq experiments
  • RNA integrity number (RIN) is a common metric used to assess RNA quality, with a score of 7 or higher generally considered acceptable
• Library preparation converts RNA into cDNA fragments with adapters attached to the ends
  • Ribosomal RNA (rRNA) depletion or poly(A) selection is used to enrich for desired RNA species (mRNA or total RNA)
  • cDNA synthesis is performed using random hexamers or oligo(dT) primers
  • Adapters contain unique barcodes for multiplexing multiple samples in a single sequencing run
• Sequencing platforms use different technologies to generate high-throughput sequencing data
  • Illumina sequencing is based on the sequencing-by-synthesis approach and is widely used for RNA-Seq
  • Pacific Biosciences (PacBio) and Oxford Nanopore Technologies offer long-read sequencing, which is useful for detecting complex isoforms and full-length transcripts
• Paired-end sequencing generates reads from both ends of a cDNA fragment, providing more accurate alignment and improved detection of splice junctions compared to single-end sequencing

Data Quality Control and Preprocessing

• Quality control (QC) is essential to ensure the reliability and reproducibility of RNA-Seq data
• FastQC is a widely used tool for assessing the quality of raw sequencing reads
  • Provides information on per-base sequence quality, GC content, adapter content, and overrepresented sequences
• Low-quality bases and adapter sequences are trimmed using tools like Trimmomatic or Cutadapt to improve alignment accuracy
• Contamination from rRNA, mtRNA, or other unwanted sources can be assessed and filtered out if necessary
• QC metrics help determine whether the data is suitable for downstream analysis or if additional preprocessing steps are required
• Preprocessing steps may include removing PCR duplicates, filtering reads based on mapping quality, and handling multimapping reads
• Normalization methods, such as reads per kilobase of transcript per million mapped reads (RPKM), fragments per kilobase of transcript per million mapped reads (FPKM), and transcripts per million (TPM), are used to account for differences in sequencing depth and gene length across samples

Alignment and Quantification Methods

• Alignment is the process of mapping sequencing reads to a reference genome or transcriptome
  • Splice-aware aligners, such as STAR, HISAT2, and TopHat2, are used to handle reads spanning splice junctions
  • Alignment parameters, such as allowed mismatches and gaps, should be optimized based on the specific dataset and research question
• Alignment quality metrics, such as the percentage of uniquely mapped reads and the distribution of reads across genomic features (exons, introns, intergenic regions), provide insights into the quality of the alignment
• Quantification estimates the expression levels of genes or transcripts based on the aligned reads
  • Count-based methods, such as HTSeq and featureCounts, count the number of reads overlapping with each gene or transcript
  • Transcript abundance estimation methods, like RSEM and Kallisto, use probabilistic models to estimate transcript-level expression and account for multi-mapping reads
  • Normalization methods, such as TPM and FPKM, are used to facilitate comparison of expression levels across samples and genes
• Pseudoalignment methods, such as Salmon and Kallisto, provide fast and accurate quantification without the need for a full alignment to the reference genome
• Isoform-level quantification can be performed using tools like RSEM and StringTie to estimate the expression of alternative splicing variants

Differential Expression Analysis

• Differential expression analysis identifies genes that are significantly up- or down-regulated between different conditions or groups
• Statistical methods, such as DESeq2, edgeR, and limma, are used to model read counts and test for differential expression
  • These methods account for biological variability and control for false positives using techniques like the Benjamini-Hochberg procedure for multiple testing correction
• Fold change and statistical significance (p-value or adjusted p-value) are used to determine the magnitude and reliability of differential expression
• Visualization techniques, such as MA plots, volcano plots, and heatmaps, help interpret and present the results of differential expression analysis
• Batch effects, caused by technical variability in sample processing or sequencing, can confound differential expression results and should be identified and corrected using methods like ComBat or SVA
• Experimental design, including the number of biological replicates and the choice of control samples, has a significant impact on the power to detect differential expression
• Differentially expressed genes can be further analyzed for functional enrichment and pathway analysis to gain biological insights

Functional Interpretation and Pathway Analysis

• Functional interpretation and pathway analysis help researchers understand the biological significance of differentially expressed genes
• Gene Ontology (GO) enrichment analysis identifies overrepresented GO terms associated with a set of genes
  • GO terms describe gene functions in three categories: biological process, molecular function, and cellular component
  • Tools like DAVID, g:Profiler, and Enrichr can be used to perform GO enrichment analysis
• Pathway analysis identifies overrepresented biological pathways and networks associated with a set of genes
  • Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome, and BioCyc are popular pathway databases
  • Tools such as GSEA, SPIA, and Ingenuity Pathway Analysis (IPA) can be used for pathway analysis
• Upstream regulator analysis predicts the activation or inhibition of transcription factors and other upstream regulators based on the expression changes of their target genes
• Integration of RNA-Seq data with other omics data, such as proteomics or metabolomics, can provide a more comprehensive understanding of biological systems
• Functional interpretation and pathway analysis results should be critically evaluated and validated using experimental methods, such as qPCR, Western blotting, or functional assays

Advanced Topics and Future Directions

• Single-cell RNA-Seq (scRNA-Seq) enables the transcriptomic profiling of individual cells, revealing cellular heterogeneity and rare cell types
  • Challenges in scRNA-Seq data analysis include data sparsity, technical noise, and batch effects
  • Computational methods for scRNA-Seq data analysis include dimensionality reduction (PCA, t-SNE, UMAP), clustering (k-means, hierarchical clustering, graph-based clustering), and trajectory inference (pseudotime analysis)
• Spatial transcriptomics combines RNA-Seq with spatial information, allowing the study of gene expression patterns in the context of tissue architecture
  • Techniques like FISSEQ, MERFISH, and 10x Visium enable the spatial mapping of transcripts
  • Computational challenges in spatial transcriptomics include data integration, normalization, and spatial pattern recognition
• Long-read sequencing technologies, such as PacBio and Oxford Nanopore, provide longer read lengths (>10kb) compared to short-read sequencing
  • Long reads enable the detection of complex isoforms, full-length transcripts, and long non-coding RNAs
  • Challenges in long-read RNA-Seq data analysis include higher error rates and lower throughput compared to short-read sequencing
• Integration of RNA-Seq with other omics data, such as genomics, epigenomics, and proteomics, enables a multi-omics approach to understanding biological systems
  • Challenges in multi-omics data integration include data heterogeneity, batch effects, and the development of appropriate computational methods
• Emerging technologies, such as in situ sequencing and in vivo RNA-Seq, offer new opportunities for studying RNA biology in its native context
  • In situ sequencing allows for the direct visualization of transcripts within intact tissues
  • In vivo RNA-Seq enables the study of RNA dynamics in living organisms