Gut Microbiome Study Methods

• Lecture focuses on methods to study the gut microbiome
• Primary methods are culturing and sequencing (16S, metagenomics)

Key Terms and Definitions for 16S and Metagenomics

• Amplicon Sequencing: A targeted sequencing approach that amplifies specific genomic regions, such as the 16S rRNA gene, before sequencing
• Shotgun Metagenomics: A sequencing method capturing all genetic material in a sample for functional analysis
• Library Preparation: The process involves fragmenting, tagging, and amplifying DNA before sequencing
• Next-Generation Sequencing (NGS): High-throughput sequencing technology used for analyzing microbial communities
• Paired-End Reads: Improve assembly accuracy by reading both ends of a DNA fragment
• Operational Taxonomic Unit (OTU): A clustering method grouping similar sequences based on a defined similarity threshold, which is typically 97%
• Amplicon Sequence Variant (ASV): Offers a more precise alternative to OTUs. Identifies unique sequences without clustering, using methods like DADA2 and Deblur
• Taxonomic Classification: Involves assigning taxonomy to sequences using reference databases such as SILVA, Greengenes, and RDP

Culture Based Methods for Gut Microbiome Study

• Counting cells using colony forming units (CFU)

• Interrogating isolates and co-cultures to understand interactions

• Measuring metabolic activity of cultured microbes

• Microbes recalcitrant to cultivation limits culture-based methods

• Less than 10% of microbes are estimated to be culturable, depending on community

• The Great Plate Count Anomaly described by Razumov in 1932 highlights the discrepancy between total cell counts and culturable cells

• Culturing can be difficult and time consuming. Culture conditions are proxies for the gut environment

• The downside of culture based methods is that microbes are not studied in their natural environment

The Plate Count Anomaly in Microbial Studies

• The number of microbes observed under a microscope exceeds the number that form colonies on standard culture media

• Many microorganisms are unculturable under typical lab conditions

• Standard plate counts significantly underestimate microbial diversity in environmental samples

• The discrepancy highlights culture-based microbial detection methods limitations

Culture Free Methods in Microbiome Research

• Molecular and sequence-based approaches determine phylogeny
• There is no need to grow organisms, can discover new and low abundance taxa
• Culture free methods rely on established databases and protocols
• Less expensive and time consuming than culture-based methods

The 16S-based approach

• DNA is extracted from a microbial community sample
• The 16S rRNA gene is amplified and sequenced
• Sequences are grouped into Operational Taxonomic Units (OTUs) based on similarity. Reference databases are used to identify OTUs
• Relative abundance to determine community composition, including organism presence, variant sequences, and SNPs

Shotgun Metagenomic Approach

• DNA extracted from microbial community sample
• Sequence community DNA
• Sequences are compared to reference genomes to determine community function
• Relative abundance of gene pathways indicate functions in the community

16S rRNA Sequencing: Advantages

• More cost-effective than metagenomics, making it suitable for large studies
• Targets conserved 16S rRNA gene, specific to bacteria and archaea
• Requires less sequencing depth as it focuses on one gene
• Presents easier data analysis due to smaller datasets facilitating manageable bioinformatics processing

16S rRNA Sequencing: Limitations

• Limited taxonomic resolution, often unable to distinguish species or strains precisely

• Provides taxonomic composition but not the functional potential of the microbiome

• Primer bias can lead to underrepresentation or missing of some bacterial groups

• It cannot detect viruses or eukaryotic microbes, because it amplifies bacterial and archaeal DNA

• 16S rRNA sequencing is best for broad overview of bacterial community composition at a lower cost

Metagenomics: Advantages

• Offers higher taxonomic resolution, identifying microbes to the species and strain level
• Provides functional insights by revealing genes, metabolic pathways, and potential microbial interactions
• Detects all microbes including bacteria, archaea, viruses, fungi, and other eukaryotic microbes
• Avoids PCR primer bias as specific primers are not needed

Metagenomics (Whole-Genome Shotgun Sequencing): Limitations

• Expensive due to higher sequencing and computational costs

• Involves Complex data analysis as advanced bioinformatics tools and high computational power are needed

• Requires more DNA, needing high-quality and sufficient microbial DNA for effective sequencing

• Host DNA contamination in host-associated samples reduce microbial reads

• Metagenomics is best when detailed taxonomic resolution and functional insights into the microbiome are needed

Timeline of Microbiome Community Study

• Sequencing advancements started with first-generation (Sanger) to second-generation (NGS) and third-generation sequencing
• The key improvements are increased speed, cost reduction, read length, and scalability

Some of NGS methods

• NGS analysis consists of:
  • DNA-seq
  • RNA-seq

What makes for a useable yardstick for evolutionary comparisons?

• Has to be universally present
• Has to have conserved and variable regions
• Has to be large enough
• Has to be minimally transferred

Ribosome Anatomy

• Eukaryotic Ribosome: 28S, 5.8S, 5S, 18S
• Prokaryotic Ribosome: 23S, 5S, 16S

The 16S rRNA of Escherichia coli Structure

• Contains 1,500 bp and 9 variable regions

Prokaryotic Ribosome Components

• Large subunit (LSU) is 50S
• Small subunit (SSU) is 30S
• Assembled ribosome 70S
• Includes 5S rRNA, 23S rRNA, 16S rRNA

16S rRNA Gene Regions

• Has variable and conserved regions, plus forward and reverse primers

First-Generation Sequencing (Sanger Sequencing)

• Developed by Frederick Sanger (1977), based on chain termination method.
• Uses dideoxynucleotides to halt DNA synthesis at specific bases
• A read length of ~500-900 base pairs.
• Limitations: High cost, slow throughput, labor-intensive.

Second-Generation Sequencing (NGS)

• Introduced high-throughput sequencing with parallel processing
• Uses platforms such as Illumina, Ion Torrent and Roche 454
• Read length is 100-300 base pairs
• Faster, cheaper, scalable, suitable for large-scale projects
• Limitations: Shorter reads, sequencing errors in homopolymer regions

Third-Generation Sequencing (Long-Read Sequencing)

• Single-molecule sequencing with real-time analysis
• Platforms: PacBio (SMRT sequencing), Oxford Nanopore
• Read length: 10,000 to 100,000 base pairs
• Long reads, better assembly, detection of complex structural variations
• Higher error rates, expensive equipment

Oxford Nanopore (Long-Read Sequencing)

• Real-time monitoring through MinKNOW
• Real-time basecalling and data assessment, plus on-demand sequencing
• There's no fixed runtime and runs can be paused

PacBio Sequencing

• Relies on ZMW wells and DNA polymerase
• DNA fragments are ligated to hairpin adapters, creating circularized templates, which get loaded into Zero-Mode Waveguides (ZMWs), tiny wells containing a single DNA polymerase at the bottom
• DNA polymerase incorporates florescently labeled nucleotides as it synthesizes the complementary strand.
• Each nucleotide has a unique florescent tag, where the camera records fluorescence events in real time
• Circularized DNA allows multiple passes over the same sequence, which improves accuracy

Comparison of Sequencing Technologies

• First-Gen (Sanger Sequencing) has ~500-1,000 bp read length, is low throughput, has high cost, and is used for small-scale sequencing.
• Second-Gen (Illumina) has ~50-300 bp read length, is high throughput, has low cost, and is used on whole genomes
• Third-Gen (PacBio, Nanopore) has ~10,000-100,000+ bp read length, is moderate to high throughput, has moderate to high cost, and is used for structural variation

Next-Gen Sequencing - Illumina

• Illumina NGS includes Library Preparation, Cluster generation, Sequencing, and Alignment and data analysis

• NGS library is prepared by fragmenting a gDNA sample and ligating specialized adapters to both fragment ends

• Library is loaded into a flow cell and the fragments are hybridized hybridized to the flow cell surface. Each bound fragment is amplified into a clonal cluster through bridge amplification

• Reagents are added including fluorescently labeled nucleotides, and the flow cell is imaged to record the emission from each cluster.

• Reads are aligned to a reference sequence using bioinformatic software

DNA extraction techniques

• Basic steps:

  • Cell Harvest
  • Cell lysis
  • Protein Removal
  • DNA binding
  • Wash
  • DNA Elution

• Microbial DNA is extracted from stool, soil, water, or tissue. First a sample is collected in a sterile container

• The sample undergoes cell lysis by breaking open bacterial cells using chemicals or physical methods, such as bead beating

• The DNA is purified to remove proteins and other unwanted material

DNA extraction techniques - possible issues

• DNA may fragment, or cells may not open.
• Protein may not hydrolyze, possibly contaminating the sample
• DNA may pass through or become to loose
• DNA may not elute

Which group is likely to be different between DNA extraction methods?

• Gram positive bacteria is likely to be different

Steps: Illumina MiSeq 16S rRNA Sequencing

• Amplification of the 16S rRNA Gene achieved using PCR

• Specific primer sequences target the V-regions (V4 or V3-V4)

• Duplicates ~ 1 billion copies for sequencing

• Spacer is a short sequence to create separation between functional DNA regions, preventing structural issues during PCR or sequencing

• Index is a unique DNA sequence added to each sample that enables pooling

• Linker is a short sequence that connects DNA fragments to help with sequencing adapter attachment

• Prepared samples are combined into a single tube inside Illumina MiSeq Machine, where prepared samples are combined into a single tube.

• Has a unique index (barcode), so different samples can be sequenced in one run, using a special flow cell.

• In the Bridge Amplification & Cluster Generation, DNA fragments bind to the flow cell and form clusters and the DNA strands attach to the flow cell surface and are duplicated into clusters

• In Sequencing by Synthesis (SBS), the machine reads DNA sequence using fluorescent signals. it adds fluorescently labeled nucleotides (A, T, G, C) one by one

• Data Output & Quality Control (QC) generates raw sequencing data checked for errors that include FASTQ files that can be assesed via FastQC

File Types

• FASTQ file contains the raw sequence data,
• The FASTA file contain the processed sequence data
• FASTA is generated and results from quality filtering and assembly of sequencing reads from a FASTQ file

De novo vs Reference Genome assembly:

• In De novo Assembly a genome is constructed without using a reference sequence
• Short sequencing reads are assembled into contigs based on sequence overlap
• The Tools for this are SPAdes, Velvet, and Canu
• Assembles genomes for novel or unsequenced species and avoids reference bias
• The challenges include requiring high sequencing depth and assembly errors

Reference Genome Assembly

• Reference Genome Assembly (Mapping Assembly) aligns sequencing reads to an existing reference genome
• Sequencing reads are aligned to a known reference genome where variations SNPs are detected
• Tools include Greengenes, RDP, Silva
• Much faster and requires lower computational resources. Provides more accurate variant calling.
• Can't assemble novel sequenced and errors if the reference genome is too distantly related

16s rRNA databases

• RDP (Ribosomal Database Project) is a comprehensive database where is provided tools for taxonomic classification as well as sequence alignment and comparison
• Contains curated 16S rRNA and supports for QIIME

16S rRNA sequencing and metagenomics for a single gut microbiome sample

Programs for Microbiome Analysis (Open Source)

• QIIME2
• Dada2
• Silva
• Phyloseq
• Ape
• Metacoder
• Mothur

Data Analysis - Differential Abundance

• LDA score (Linear Discriminant Analysis score) is used to identify significantly different taxa between group

Correlation analysis data outcomes

• Useful for:
  • Veillonella
  • Megamonas
  • Dialister
  • Ruminococcus
  • Faecalibacterium Etc

Data Analysis - Community Distance

• Beta diversity is when community is a vector of abundances
• Must consider ranged distances 0 to 1
• The distance to the self must be 0

Distance Spectrum

• Can be measured with a categorical or phylogenetic

Distance Spectrum Jaccard

• Measures community similarity based on presence/absence of species
• Range: 0 (no shared species) to 1 (identical communities)

Metagenomics

• Considers all: "Who's there?", "What are they doing?" and "What does it all mean?"
• We don't always do this because
  • its expensive
  • not always needed
  • possible low abundance microbes

Sequencing Sample Bias

• Issues:
  • Inadequate sampling
  • Change distribution via storage
  • Different DNA recovery can change strains

Fecal sample as proxy for gut environment

• Issues:
  • the difference between gut sections
  • the difference between lumen and mucosa
  • and sampling logistics