👻Intro to Computational Biology Unit 7 – Gene Expression Analysis

Key Concepts in Gene Expression

• Gene expression process converts genetic information encoded in DNA into functional gene products (proteins or non-coding RNAs)
• Involves two main steps: transcription (DNA to RNA) and translation (RNA to protein)
• Tightly regulated at multiple levels (transcriptional, post-transcriptional, translational, and post-translational)
• Differential gene expression drives cellular differentiation and specialization
  • Allows cells with identical genomes to have distinct functions (neurons vs. muscle cells)
• Gene expression patterns change in response to environmental stimuli, developmental stages, and disease states
• Studying gene expression provides insights into cellular functions, regulatory networks, and disease mechanisms
• High-throughput technologies (microarrays, RNA-seq) enable genome-wide expression profiling

DNA to RNA: Transcription Basics

• Transcription initiates gene expression by synthesizing RNA from a DNA template
• Carried out by RNA polymerase enzymes (RNA polymerase II for most protein-coding genes)
• Transcription factors bind specific DNA sequences to recruit RNA polymerase and regulate transcription initiation
• Consists of three main stages: initiation, elongation, and termination
  • Initiation: RNA polymerase binds promoter region and separates DNA strands
  • Elongation: RNA polymerase moves along the template strand, synthesizing complementary RNA
  • Termination: RNA polymerase releases the newly synthesized RNA and dissociates from DNA
• Eukaryotic transcripts undergo post-transcriptional modifications (5' capping, splicing, 3' polyadenylation)
• Alternative splicing generates multiple mRNA isoforms from a single gene, increasing proteome diversity

RNA to Protein: Translation Overview

• Translation converts the genetic information in mRNA into a polypeptide chain
• Occurs in the cytoplasm on ribosomes, large RNA-protein complexes
• Genetic code specifies the correspondence between mRNA codons and amino acids
  • Each codon (triplet of nucleotides) encodes a specific amino acid or stop signal
• tRNAs act as adaptor molecules, carrying amino acids to the ribosome and recognizing codons via anticodon sequences
• Consists of three main stages: initiation, elongation, and termination
  • Initiation: Ribosomal subunits assemble on the mRNA with the help of initiation factors
  • Elongation: tRNAs deliver amino acids, which are linked together by peptide bonds
  • Termination: Release factors recognize stop codons and trigger polypeptide release
• Post-translational modifications (folding, cleavage, chemical modifications) generate mature, functional proteins

Gene Regulation Mechanisms

• Gene regulation controls the timing, location, and level of gene expression
• Transcriptional regulation involves controlling the rate of transcription initiation
  • Transcription factors bind regulatory DNA sequences (promoters, enhancers, silencers) to activate or repress transcription
  • Chromatin structure and epigenetic modifications (DNA methylation, histone modifications) influence gene accessibility
• Post-transcriptional regulation targets mRNA stability, localization, and translation efficiency
  • MicroRNAs (miRNAs) and RNA-binding proteins (RBPs) are key regulators at this level
• Translational regulation controls the rate of protein synthesis from mRNA
  • Includes mechanisms like ribosome recruitment, start codon recognition, and translational repression
• Post-translational regulation modifies protein activity, stability, and localization
  • Phosphorylation, ubiquitination, and other modifications can alter protein function and half-life
• Feedback loops and regulatory networks enable precise control and coordination of gene expression

High-Throughput Sequencing Technologies

• High-throughput sequencing (HTS) technologies enable massive parallel sequencing of DNA or RNA
• RNA sequencing (RNA-seq) is widely used for transcriptome profiling and gene expression analysis
  • Provides a quantitative measure of transcript abundance across the genome
• Involves converting RNA to cDNA, fragmentation, adapter ligation, and sequencing
• Generates millions of short reads that are mapped back to a reference genome or transcriptome
• Offers several advantages over microarrays: higher sensitivity, dynamic range, and ability to detect novel transcripts
• Single-cell RNA-seq (scRNA-seq) allows expression profiling at the individual cell level
  • Captures cell-to-cell heterogeneity and identifies rare cell types
• Other HTS applications: ChIP-seq (protein-DNA interactions), ATAC-seq (chromatin accessibility), ribosome profiling (translation)

Computational Tools for Expression Analysis

• Quality control: Assessing read quality, trimming adapters, and filtering low-quality reads (FastQC, Trimmomatic)
• Read alignment: Mapping reads to a reference genome or transcriptome (STAR, HISAT2, Bowtie2)
• Quantification: Estimating transcript or gene abundance from aligned reads (featureCounts, HTSeq, Kallisto)
• Normalization: Adjusting for differences in library size and composition (TPM, RPKM, DESeq2, edgeR)
• Differential expression analysis: Identifying genes with significant expression changes between conditions (DESeq2, edgeR, limma)
• Clustering: Grouping samples or genes based on expression patterns (hierarchical clustering, k-means, t-SNE)
• Pathway and gene set enrichment analysis: Identifying overrepresented biological functions or pathways (GSEA, GO enrichment)
• Data integration: Combining expression data with other omics data types (ChIP-seq, ATAC-seq, proteomics)

Statistical Methods in Gene Expression Studies

• Normalization methods account for technical biases and enable fair comparisons across samples
  • Common methods: TPM (transcripts per million), RPKM (reads per kilobase per million), DESeq2, edgeR
• Differential expression analysis identifies genes with significant expression changes between conditions
  • Based on statistical tests (e.g., Wald test, likelihood ratio test) and fold change thresholds
  • Multiple testing correction controls false positives (FDR, Bonferroni)
• Clustering algorithms group samples or genes based on expression similarity
  • Hierarchical clustering: Builds a tree-like structure based on pairwise distances
  • K-means clustering: Partitions data into a predefined number of clusters
• Principal component analysis (PCA) reduces data dimensionality and visualizes major sources of variation
• Gene set enrichment analysis (GSEA) assesses the enrichment of predefined gene sets in ranked gene lists
• Machine learning methods (e.g., random forests, support vector machines) can predict sample classes or outcomes based on expression signatures

Interpreting and Visualizing Expression Data

• Heatmaps display expression levels across samples and genes using color gradients
  • Rows (genes) and columns (samples) are often clustered to reveal patterns
• Volcano plots combine statistical significance ($-log_{10}(p-value)$) and magnitude of change ($log_2(fold change)$)
  • Helps identify genes with large and significant expression changes
• MA plots compare expression levels ($log_2(mean expression)$) and fold changes ($log_2(fold change)$)
  • Useful for assessing differential expression analysis results and identifying outliers
• Principal component analysis (PCA) plots visualize sample relationships in reduced dimensional space
• Gene set enrichment plots (e.g., GSEA enrichment plot, GO term bar plots) summarize the enrichment of biological functions or pathways
• Network diagrams depict gene-gene interactions, co-expression relationships, or regulatory networks
• Interactive visualization tools (e.g., Shiny apps, Plotly) enable dynamic exploration of expression data

Applications in Research and Medicine

• Identifying disease biomarkers: Expression signatures associated with disease states can serve as diagnostic or prognostic markers
• Drug discovery and development: Expression profiling can identify drug targets, assess drug efficacy, and predict side effects
• Studying cellular differentiation and development: Expression dynamics during cell fate transitions provide insights into developmental processes
• Characterizing tumor heterogeneity: Single-cell expression profiling reveals subpopulations within tumors with distinct properties
• Investigating host-pathogen interactions: Expression changes in host cells upon infection shed light on immune responses and pathogenesis
• Precision medicine: Expression-based patient stratification can guide personalized treatment decisions
• Functional genomics: Integrating expression data with other omics data types to understand gene functions and regulatory networks
• Comparative transcriptomics: Comparing expression patterns across species to study evolution and conservation of biological processes