Day 4: Genome Assembly Essentials: QC, Identification, and assembly¶
Date: September 4, 2025
Duration: 09:00-13:00 CAT
Focus: Genome assembly and assessment
Overview¶
Day 4 builds on Day 3 by starting with a recap on quality control, followed by genome assembly and assessment of assemblies to ensure.
Learning Objectives¶
By the end of Day 4, you will be able to:
- Execute de novo genome assembly using SPAdes or other assemblers
- Assess assembly quality using QUAST and other metrics
Schedule¶
| Time (CAT) | Topic | Links | Trainer |
|---|---|---|---|
| 09:00 | Recap: Quality checking and control, and species identification | Practical | Arash Iranzadeh |
| 10:30 | Genome assembly, quality assessment | Notes. | Ephifania Geza |
| 11:00 | Break | ||
| 11:30 | Genome assembly, quality assessment: Continuation | Practical | Ephifania Geza |
Key Topics¶
1. Genome assembly and assessing contigs¶
- De novo assembly algorithms and approaches
- Short-read vs long-read assembly strategies
- Assembly quality metrics and interpretation
- Assembly polishing and gap filling
Datasets Used¶
V. cholerae Outbreak Collection¶
- Source: Multi-country cholera outbreak (2019)
- Samples: 25 clinical isolates + environmental samples
- Timespan: 8-month outbreak period
- Geographic: Three countries, coastal regions
- Epidemiological data: Case demographics, travel history
Tools Introduced¶
Pangenome Analysis¶
- SPAdes - De novo genome assembler
- Unicycler - Hybrid assembly pipeline
- Flye - Long-read assembly
- QUAST - Assembly quality assessment
Annotation Tools¶
- Prokka - Automated prokaryotic annotation
- RAST - Rapid Annotation using Subsystem Technology
- NCBI PGAP - Prokaryotic Genome Annotation Pipeline
-
Bakta - Rapid bacterial genome annotation
-
Panaroo - Pangenome pipeline
- Roary - Rapid large-scale prokaryote pangenome analysis
- PPanGGOLiN - Depicting microbial diversity via pangenomes
Phylogenetic Analysis¶
- IQ-TREE - Maximum likelihood phylogenetic inference
- RAxML - Randomized Axelerated Maximum Likelihood
- FastTree - Approximately maximum-likelihood trees
SNP Analysis¶
- Snippy - Rapid haploid variant calling
- ParSNP - Rapid core genome SNP typing
- Gubbins - Recombination detection in bacterial genomes
Hands-on Exercises¶
Exercise 1: Genome Assembly and Quality Assessment (60 minutes)¶
Assemble bacterial genomes and evaluate assembly quality.
# De novo assembly with SPAdes
spades.py --careful -1 sample_R1.fastq.gz -2 sample_R2.fastq.gz -o assembly_output/
# Assembly quality assessment
quast.py assembly_output/scaffolds.fasta -o quast_results/
# Check assembly statistics
assembly-stats assembly_output/scaffolds.fasta
### Exercise 1: Pangenome Analysis (60 minutes)
Analyze the core and accessory genome of *V. cholerae* outbreak strains.
```bash
# Run Panaroo pangenome analysis
panaroo -i *.gff -o panaroo_output --clean-mode strict
# Visualize results
python3 scripts/visualize_pangenome.py panaroo_output/
Exercise 2: Phylogenetic Tree Construction (60 minutes)¶
Build maximum likelihood trees from core genome alignments.
# Generate core genome alignment
snippy-core --ref reference.gbk snippy_output/*
# Build phylogenetic tree
iqtree -s core_alignment.aln -m GTR+G -bb 1000 -nt 4
# Visualize tree
figtree core_alignment.aln.treefile
Exercise 3: Outbreak Investigation (45 minutes)¶
Integrate genomic and epidemiological data to investigate transmission.
# Calculate pairwise SNP distances
snp-dists core_alignment.aln > snp_distances.tsv
# Identify transmission clusters
cluster_analysis.py --snp-threshold 10 --epi-data metadata.csv
Key Concepts¶
Assembly Quality Metrics¶
| Metric | Good Assembly | Poor Assembly | Action |
|---|---|---|---|
| N50 | >50 kb | <10 kb | Optimize parameters |
| Contigs | <100 | >500 | Check contamination |
| Genome size | Expected ±10% | >20% difference | Review input data |
| Coverage | >50x | <20x | Sequence more |
Pangenome Structure¶
- Core genome: Genes present in all strains (housekeeping functions)
- Accessory genome: Variable genes (adaptation, virulence, resistance)
- Singleton genes: Present in single strain only
- Shell genes: Present in several but not all strains
Phylogenetic Inference¶
- Substitution models: GTR, HKY, JC69 for nucleotide evolution
- Rate heterogeneity: Gamma distribution for variable sites
- Bootstrap support: Statistical confidence in tree topology
- Branch lengths: Evolutionary distance (substitutions per site)
SNP Thresholds for Outbreak Investigation¶
| Pathogen | SNP Threshold | Timeframe | Context |
|---|---|---|---|
| M. tuberculosis | 0-5 SNPs | Recent transmission | Same household |
| S. Typhi | 0-20 SNPs | Outbreak cluster | Weeks to months |
| V. cholerae | 0-10 SNPs | Epidemic spread | Days to weeks |
| E. coli O157 | 0-15 SNPs | Foodborne outbreak | Days |
Advanced Topics¶
Molecular Dating¶
- Tip dating: Using collection dates for molecular clock
- Bayesian methods: BEAST, MrBayes for time-resolved phylogenies
- Substitution rates: Pathogen-specific evolutionary rates
Network Analysis¶
- Minimum spanning trees: Alternative to bifurcating phylogenies
- Median-joining networks: Visualization of reticulate evolution
- Transmission networks: Direct transmission inference
Assessment Activities¶
Individual Analysis¶
- Generate pangenome summary statistics
- Construct phylogenetic tree with bootstrap support
- Calculate SNP distances between outbreak isolates
- Identify transmission clusters based on genomic data
Group Discussion¶
- Compare assembly strategies and results
Common Challenges¶
- Interpret pangenome diversity in outbreak context
- Evaluate phylogenetic support for transmission hypotheses
- Discuss integration of genomic and epidemiological evidence
- Consider limitations of genomic outbreak investigation
Common Challenges¶
Low Phylogenetic Resolution¶
# Try different substitution models
iqtree -s alignment.aln -m MFP -bb 1000 # Model selection
# Use more informative sites only
iqtree -s alignment.aln -m GTR+G -bb 1000 --rate
Recombination Issues¶
# Detect and remove recombinant regions
gubbins alignment.aln
# Use recombination-free alignment
iqtree -s alignment.filtered_polymorphic_sites.fasta
Missing Epidemiological Links¶
# Lower SNP threshold for exploration
cluster_analysis.py --snp-threshold 20 --epi-data metadata.csv
# Consider longer transmission chains
transmission_chains.py --max-generations 3
Resources¶
Resources¶
Assembly Resources¶
Key Publications¶
- Page et al. (2015). Roary: rapid large-scale prokaryote pangenome analysis
- Tonkin-Hill et al. (2020). Producing polished prokaryotic pangenomes
- Croucher et al. (2015). Rapid phylogenetic analysis of bacterial genomes
Software Documentation¶
Online Resources¶
- Microreact - Visualization platform
- iTOL - Interactive tree visualization
- FigTree - Tree viewing software
Looking Ahead¶
Day 5 Preview: Metagenomics analysis including: - Microbiome profiling from clinical samples - Pathogen detection in complex communities - Functional analysis of metagenomes - Association with host health outcomes
Homework (Optional)¶
- Analyze pangenome diversity in additional pathogen datasets
- Experiment with different phylogenetic methods and compare results
- Read case studies of genomic outbreak investigations
- Practice tree interpretation with published examples
Key Takeaway: Genomic analysis provides unprecedented resolution for understanding pathogen evolution and transmission, but interpretation requires careful integration with epidemiological context and understanding of method limitations.