Genome, Transcriptome Assembly, and Annotations

Genome Transcriptome Assembly and Annotations

Zoonemia Project

• A project involving the assembly of genomes from 240 different mammals.
• The assembled genomes, along with human genome data, are available for download.
• A key publication from the Zoonemia Consortium is essential reading and will be included in exam questions.

Genome Assembly

• Genome assembly involves taking sequenced DNA fragments and putting them together to create a comprehensive genome sequence.
• The process is complicated by the presence of repetitive elements, which can make up a significant portion (e.g., 50% or more) of the genome.
• Gaps may remain in the initial assembly due to the difficulty in aligning repetitive regions.
• The human genome was the first to be completed with telomere-to-telomere assembly, a process that took approximately 20 years.

UCSC Genome Browser

• The UCSC Genome Browser provides access to assembled genome sequences.
• The human genome assembly was completed in January 2022, starting from its initial release in September 2001. The "T2T" designation means telomere to telomere, indicating a complete sequence from one end of the chromosome to the other.

Assembly and Annotation

• After sequencing, the DNA fragments are assembled into contigs, and then further into scaffolds.
• Annotations are then added to the assembled genome. These annotations include structural annotations (locating genes) and functional annotations (determining the function of genes).
• Gene Ontology (GO) and KEGG pathways are used to assign meaning to the genome, indicating the functional importance of different genomic regions.

Assembly Steps

• The genome assembly process involves several steps:
  • Sequencing reads obtained from Illumina, PacBio, or other platforms.
  • Assembly of reads into contigs.
  • Arrangement of contigs into scaffolds.
• Gap filling is a challenging and expensive process, as demonstrated by the 20 years it took to complete the human genome.
• The final step is assembling the complete genome sequence.

Types of Assembly

• Reference-based assembly:
  • Involves using an existing reference genome to guide the assembly process.
  • The chimpanzee genome was assembled using the human genome as a reference due to their high similarity.
• De Novo assembly:
  • Involves assembling a genome without a reference.
  • This approach is used for organisms without a known, closely related reference genome.
  • Examples include endemic species like Salmo trutta. Anatolian leopard (Pardus pardus) is another species where de novo assembly is essential.

De Novo Assembly Process

• The de novo assembly process includes:
  • Sequencing.
  • Quality control and trimming.
  • Removal of contamination.
  • K-mer counting to estimate genome size.
  • Error correction.
  • De novo assembly using algorithms like Velvet (for DNA) and Trinity (for RNA).
  • Annotation.

Sequencing Technologies

• Combining short-read (Illumina) and long-read (PacBio or Oxford Nanopore) sequencing is optimal.
• Illumina provides high-quality data at a low cost, while long-read sequencing helps to fill gaps and resolve repetitive regions.
• Different algorithms are used for de novo assembly, such as the overlap layout algorithm (used by SAIC) and the de Bruijn graph algorithm (used by Velvet).

Assembly Algorithms

• Key algorithms include:
  • Overlap-Layout-Consensus (OLC) Algorithm: Reads are assembled based on overlapping regions, useful but struggles with repetitive parts.
  • De Bruijn Graph Algorithm: Sequences are broken into smaller k-mers (short nucleotide sequences), which are then assembled into a graph. Trinity is a successful program that uses this algorithm.

N50 Value

• The N50 value is a statistical measure used to assess the quality of a genome assembly.
• It represents the minimum contig length needed to cover 50% of the genome.
• A higher N50 value indicates a better assembly.

Key Factors for Genome Assembly

• Important factors for genome assembly include:
  • Genome size.
  • Heterozygosity.
  • GC content (high GC content can cause problems).
  • High-quality DNA.
  • Appropriate sequencing technology (combination of Illumina and PacBio).
  • Computational resources.

Genome Assembly Workflow

• The general workflow for genome assembly involves:
  • Quality control.
  • Trimming.
  • Assembly.
  • Validation.

Genome Annotation

• Genome annotation is a critical step after assembly.
• It involves identifying the locations of genes (structural annotation) and determining their functions (functional annotation).
• Only a small percentage of the genome (e.g., 20-25%) is typically functionally annotated.

Annotation Types

• Structural Annotation: Identifies gene locations, exon-intron structure, and other genomic features.
• Functional Annotation: Assigns functions to genes and other genomic elements.

In Silico vs. Experimental Annotation

• Annotation can be done in silico (computationally) or through experimental validation.
• In silico annotations are predictions based on sequence analysis.
• Experimental validation is needed to confirm the accuracy of these predictions.
• Molecular biologists play a crucial role in experimentally validating computational predictions.

Transcriptomics

• Transcriptomics involves studying the transcriptome, which is the complete set of RNA transcripts in a cell or organism.
• This includes mRNA, alternative splicing variants, and non-coding RNAs.
• RNA sequencing (RNA-seq) is used to analyze the transcriptome.
• Transcriptome assembly can be done de novo (using Trinity) or with reference to a genome (using cufflinks).
• The goal is to identify different isoforms and splicing architectures.

Transcriptome Assembly

• Transcriptome assembly aims to identify different RNA isoforms and splicing variations.
• The Trinity method, developed by the Broad Institute and the Hebrew University of Jerusalem, utilizes a De Bruijn graph approach.

Trinity Strategies

• Trinity uses three main strategies:
  • Inchworm: Creates a linear sequence of transcripts with a greedy approach.
  • Chrysalis: Clusters Inchworm contigs into connected components.
  • Butterfly: Processes the graphs to resolve alternative splicing and isoform variations.

RNA Sequencing and Assembly

• Paired-end RNA sequencing is crucial for transcriptome assembly.
• The assembly can be reference-based (if a similar genome is available) or de novo.
• De novo assembly helps identify alternative splicing events and transcript variations.

Read Depth and Saturation

• The required number of reads depends on the genome and transcriptome complexity.
• For yeast (Saccharomyces), approximately 45 million reads may be sufficient, while for mouse, at least 90 million reads may be needed to reach saturation.

Assessing Assembly Correctness

• Statistical tools are used to assess the correctness of assemblies.
• The N50 value and tools like H750 can be used to evaluate assembly quality.

Comparative Transcriptomics

• Assembled transcriptomes can be compared to study gene expression differences.
• Expression data provides additional information on phenotypic and functional differences between species.
• A publication by Allen Wilson and Marie-Claire King demonstrated the expression differences can give new information about species relationships.

Brain Transcriptomics Comparison

• Comparative transcriptomics can reveal differences in gene usage across species.
• For example, genes in the hippocampus region of the brain show different expression ratios in humans, mice, and pigs.

Computation Biology Pioneers

• Eugene Myers is recognized as one of the founders of computational biology. He developed the first assembly program.
• David Haussler and Jim Kent are also pioneers, known for the UCSC Genome Browser.