🧬Computational Genomics Unit 6 – Regulatory Genomics & Epigenomics

Key Concepts and Definitions

• Regulatory genomics studies how gene expression is controlled by various regulatory elements and mechanisms
• Epigenomics focuses on heritable changes in gene expression that do not involve alterations to the DNA sequence itself
• Transcription factors (TFs) are proteins that bind to specific DNA sequences and regulate transcription of target genes
• Enhancers are distal regulatory elements that positively regulate gene expression by interacting with promoters through DNA looping
  • Can be located upstream or downstream of the genes they regulate (e.g., the sonic hedgehog enhancer located ~1 Mb upstream of the SHH gene)
• Silencers are regulatory elements that negatively regulate gene expression by recruiting repressive factors
• Insulators are boundary elements that prevent inappropriate interactions between neighboring chromatin domains
• Chromatin accessibility refers to the degree to which DNA is accessible to TFs and other regulatory proteins
  • Open chromatin regions are associated with active gene expression, while closed chromatin is associated with gene repression

Regulatory Elements in the Genome

• Promoters are located near the transcription start site (TSS) and contain binding sites for RNA polymerase and general TFs
• Proximal promoters are located immediately upstream of the TSS and contain core promoter elements (TATA box, initiator, downstream promoter element)
• Distal promoters are located further upstream and contain additional regulatory elements (CpG islands, upstream activating sequences)
• Enhancers can be located far from their target genes and interact with promoters through DNA looping mediated by cohesin and mediator complexes
  • Super-enhancers are clusters of enhancers that drive high levels of gene expression in cell type-specific manner (e.g., the α-globin super-enhancer in erythroid cells)
• Silencers can be located near or far from their target genes and recruit repressive factors (histone deacetylases, polycomb group proteins)
• Insulators prevent enhancer-promoter interactions and chromatin spreading by forming loops and interacting with nuclear lamina
  • CTCF is a key insulator-binding protein that mediates chromatin looping and TAD formation

Epigenetic Modifications and Mechanisms

• DNA methylation involves the addition of methyl groups to cytosine residues, primarily at CpG dinucleotides
  • Methylation of promoter CpG islands is associated with gene silencing, while methylation of gene bodies is associated with active transcription
• Histone modifications include acetylation, methylation, phosphorylation, and ubiquitination of histone tails
  • H3K4me3 is associated with active promoters, while H3K27me3 is associated with repressed promoters and enhancers
  • H3K27ac is associated with active enhancers and distinguishes them from poised enhancers marked by H3K4me1 alone
• Chromatin remodeling involves the ATP-dependent alteration of nucleosome positioning and composition by remodeling complexes (SWI/SNF, ISWI, CHD, INO80)
• Non-coding RNAs (ncRNAs) can regulate gene expression through various mechanisms
  • Long non-coding RNAs (lncRNAs) can recruit chromatin-modifying complexes, act as enhancer RNAs, or serve as scaffolds for protein complexes (e.g., XIST lncRNA in X chromosome inactivation)
  • microRNAs (miRNAs) can post-transcriptionally repress gene expression by targeting mRNAs for degradation or translational repression

Experimental Techniques in Regulatory Genomics

• Chromatin immunoprecipitation followed by sequencing (ChIP-seq) allows genome-wide mapping of protein-DNA interactions
  • Used to identify binding sites of TFs, histone modifications, and chromatin-associated proteins
  • Requires antibodies specific to the protein of interest and sufficient cell numbers for robust signal
• DNase-seq and ATAC-seq identify regions of open chromatin by digesting accessible DNA with DNase I or transposase, respectively
  • Open chromatin regions are indicative of active regulatory elements (promoters, enhancers) and TF binding sites
• Bisulfite sequencing determines DNA methylation patterns by converting unmethylated cytosines to uracil while leaving methylated cytosines unchanged
  • Whole-genome bisulfite sequencing (WGBS) provides single-base resolution methylation profiles, but is costly and requires high coverage
  • Reduced representation bisulfite sequencing (RRBS) focuses on CpG-rich regions and is more cost-effective
• Chromosome conformation capture (3C) techniques detect long-range chromatin interactions
  • Hi-C provides genome-wide interaction maps at kilobase to megabase resolution, revealing topologically associating domains (TADs) and chromatin loops
  • Chromatin interaction analysis by paired-end tag sequencing (ChIA-PET) identifies interactions mediated by specific proteins (e.g., CTCF, RNA polymerase II)

Computational Methods for Epigenomic Analysis

• Peak calling identifies enriched regions in ChIP-seq and accessibility data using algorithms (MACS2, HOMER, F-Seq) that compare signal to background
• Differential analysis identifies regions with significant differences in signal intensity between conditions or cell types using tools like DiffBind and DESeq2
• Chromatin state annotation segments the genome into distinct states (active promoter, strong enhancer, repressed, etc.) based on combinatorial histone modification patterns using hidden Markov models (ChromHMM, Segway)
• Motif analysis identifies enriched DNA sequence motifs in regulatory regions using de novo discovery tools (MEME, DREME) or known motif scanning (FIMO, HOMER)
  • Motif instances can be used to infer TF binding and construct regulatory networks
• Integrative analysis combines multiple data types (e.g., ChIP-seq, RNA-seq, Hi-C) to gain insights into regulatory mechanisms and predict functional effects of genetic variants
  • Tools like GREAT and RegulomeDB annotate variants with regulatory information and predict their impact on gene expression
• Machine learning approaches (e.g., deep learning) are increasingly used to predict regulatory elements, chromatin states, and gene expression from DNA sequence and epigenomic data

Regulatory Networks and Gene Expression

• Gene regulatory networks (GRNs) describe the complex interactions between TFs and their target genes that control cell type-specific gene expression programs
  • Can be constructed using TF binding data, gene expression data, and computational inference methods (e.g., ARACNE, GENIE3)
• Transcriptional regulation involves the interplay between TFs, co-factors, and chromatin state to control the initiation and rate of transcription
  • General TFs (TFIIA, TFIIB, TFIID, TFIIE, TFIIF, TFIIH) assemble at the promoter to form the pre-initiation complex and recruit RNA polymerase II
  • Sequence-specific TFs bind to enhancers and promoters to activate or repress transcription by recruiting co-activators or co-repressors
• Post-transcriptional regulation modulates gene expression through mRNA processing, stability, and translation
  • Alternative splicing generates transcript isoforms with different functions or stability
  • miRNAs and RNA-binding proteins (RBPs) regulate mRNA stability and translation efficiency
• Feedback loops and feed-forward loops are common motifs in GRNs that enable precise control of gene expression dynamics
  • Negative feedback loops confer robustness and homeostasis, while positive feedback loops amplify signals and generate switch-like responses

Applications in Health and Disease

• Genome-wide association studies (GWAS) have identified numerous disease-associated variants, many of which lie in non-coding regulatory regions
  • Integrating GWAS data with epigenomic data can help prioritize causal variants and elucidate their functional effects on gene regulation
• Epigenetic alterations are implicated in various diseases, including cancer, neurodevelopmental disorders, and autoimmune diseases
  • DNA methylation changes and aberrant histone modifications can lead to altered gene expression and disease progression
  • Epigenetic drugs targeting DNA methyltransferases (DNMTs) and histone deacetylases (HDACs) are used in cancer treatment (e.g., azacitidine, vorinostat)
• Personalized medicine approaches leverage epigenomic data to stratify patients, predict drug responses, and develop targeted therapies
  • Epigenetic biomarkers can be used for early detection, prognosis, and treatment selection in various diseases
• Epigenetic inheritance and transgenerational effects are areas of active research
  • Epigenetic modifications can be inherited across generations and may contribute to disease risk and evolutionary adaptation
  • Environmental factors (diet, stress, toxins) can induce epigenetic changes that affect offspring health and development

Emerging Trends and Future Directions

• Single-cell epigenomics technologies (scRNA-seq, scATAC-seq, scBS-seq) enable the study of epigenetic heterogeneity and cell type-specific regulatory landscapes
  • Help identify rare cell types, developmental trajectories, and epigenetic states associated with disease
• Spatial epigenomics methods (e.g., spatially resolved ChIP-seq, spatial transcriptomics) provide information on the spatial organization of regulatory elements and gene expression in tissues
• CRISPR-based epigenome editing tools allow targeted manipulation of DNA methylation and histone modifications at specific loci
  • Used to dissect the functional roles of epigenetic modifications and regulatory elements in gene regulation and disease
• Integration of multi-omics data (epigenomics, transcriptomics, proteomics, metabolomics) using systems biology approaches will provide a more comprehensive understanding of gene regulation and its impact on cellular phenotypes
• Machine learning and artificial intelligence will play an increasingly important role in analyzing and interpreting large-scale epigenomic datasets
  • Deep learning models can predict epigenetic states, gene expression, and disease outcomes from DNA sequence and other features
• Comparative epigenomics across species will shed light on the evolution of regulatory mechanisms and their role in adaptation and speciation
• Epigenetic clocks based on DNA methylation patterns can predict biological age and are being developed as biomarkers of aging and disease risk