🧬 Overview: From Bacteria to Bioinformatics

The lecture focuses on the 16S rRNA gene — the genetic “fingerprint” used to identify bacteria.

• Conserved regions: Allow universal primer binding.
• Variable regions: Distinguish between species (evolutionary hotspots). Workflow:

1. Sampling bacteria from an environment.
2. Extracting total DNA and targeting the 16S gene.
3. Amplifying with PCR to create amplicons.
4. Sequencing and matching against databases. ➡️ End result: A table of taxa (bacterial names) and their relative abundances.

🧪 Sampling Theory — “Garbage In, Garbage Out”

If your sample is unrepresentative, your entire analysis is meaningless. Key considerations:

• Experimental design: Define what question you’re asking.
• Replication:
  • Biological replicates = independent samples.
  • Technical replicates = repeated measurements.
• Representativeness: Choose sampling sites carefully.

💧 Example (Wastewater Treatment Plant): Multiple bioreactors were sampled (B1–B5).

• Samples near the inflow were inconsistent due to unsteady mixing.
• Samples from the well-mixed zones at the back were consistent ✅ 👉 Always sample where conditions are stable and mixed.

🧊 Storage & Handling

Samples can degrade!

• Short-term: keep at ~4 °C temporarily.
• Long-term: freeze at −20 °C or lower.
• Test whether storage affects results. Also consider removing environmental DNA (free DNA from dead cells). 💡 Use PMA/TMA treatment to bind and remove free DNA — ensures only living bacteria are analyzed.

🔬 DNA Extraction Pitfalls

Key factor: Bead-beating intensity (cell lysis).

• Too weak → low DNA yield, miss tough bacteria (e.g. Actinobacteria).
• Too strong → fragment DNA, reduce quality. ✅ Find a compromise — good yield without over-shearing. 💡 Always check fragment quality (e.g. by gel electrophoresis).

🧫 Validation — Independent Checks

Use FISH (Fluorescence In Situ Hybridization) 🧪

• Targets RNA in intact cells with fluorescent probes.
• Confirms which bacteria are truly present and alive. Comparing 16S sequencing vs. FISH strengthens conclusions.

🧬 Primer Selection & PCR Conditions

Primers determine which bacteria you can detect.

• Common sets:
  • V3–V4 (515F–806R) → “universal” coverage.
  • V1–V3 → better for wastewater systems.
  • Full-length (V1–V8) for long-read sequencing (e.g. Nanopore). ⚠️ Many bacteria lack the V9 region, so V1–V8 is usually safer.

Check primer coverage in silico using tools like SILVA TestPrime or MiDAS databases. 💡 Example: V3–V4 underestimates Chloroflexi — a key phylum in wastewater plants.

PCR optimization:

• Vary annealing temperature (e.g. 52 °C often problematic).
• Optimize cycles and DNA amount per reaction. Follow manufacturer’s guide, then validate experimentally.

🧩 Multiple 16S Copies & Abundance Bias

Many bacteria have more than one copy of the 16S gene.

• Proteobacteria may have 1–16 copies!
• Therefore, sequencing counts ≠ true abundance. Combine 16S data with genomic copy-number correction or FISH counts.

⚙️ PCR & Sequencing Errors

Sources of bias:

• Amplification bias (some templates amplify better).
• PCR errors (mutations, chimeras).
• Sequencing errors (base calling, insufficient depth).
• Incomplete primer coverage (missing taxa).

ASVs vs OTUs

🧠 Tip: Use ASVs when possible for reproducible, fine-scale taxonomy.

🧠 Databases & Taxonomic Assignment

Databases (SILVA, MiDAS, etc.) are used to match 16S reads.

• Can be incomplete or contain low-quality sequences.
• Result: many reads remain “unknown.”

If unclassified:

1. BLAST in another database.
2. If still no match → potential novel species 🧫
   • Validate by genome sequencing or FISH probes.

Use ecosystem-specific databases (like MiDAS for wastewater). They provide:

• Higher-quality reference sequences.
• Better taxonomic resolution.
• Placeholder names (e.g. MiDAS_G123) for reproducibility. 📘 “Candidatus” means known only from DNA, never cultured.

📊 Interpretation Limits

The 16S dataset gives relative abundance, not true absolute abundance.

• Reflects presence trends, not exact counts.
• For precise quantification, combine with:
  • FISH
  • Metagenomics
  • Isolations
  • qPCR

💥 Practical Applications in Wastewater Systems

Even with all limitations, 16S sequencing is powerful when applied correctly.

🧼 1. Foaming Problems

• Caused by overgrowth of filamentous bacteria like Gordonia.
• Identification is vital — different foams need different chemical treatments. 💰 Misidentification = wasted money and unresolved foaming.

🌍 2. The MiDAS Project

MiDAS (Microbial Database for Activated Sludge)

• Samples taken globally 🌎
• Provides ecosystem-specific taxonomy + function links.
• Enables:
  • Linking taxonomy → function.
  • Understanding process performance.
  • Designing FISH probes for visualization.

Each wastewater plant has a unique microbial fingerprint. Differences arise from:

• Temperature
• Industrial input
• Operation style
• Mixing and aeration patterns

🔁 3. “Fecal Transplant” Analogy

Researchers transplanted microbial communities from a healthy plant to a failing one.

• Initially, the recipient community resembled the donor.
• After ~48 days, it reverted to its original state. 🧠 Lesson: environmental conditions (temperature, inflow, dispersal) dominate over microbial introduction.

💻 4. Nanopore Field Sequencing

Portable devices allow on-site DNA sequencing in real time. Potential: biological sensors for monitoring plant health.

📈 5. Predictive Microbial Ecology

Long-term monitoring enables machine learning predictions.

• Using Graph Neural Networks, they can forecast bacterial blooms (e.g. Microthrix parvicella) up to 2 months in advance.
• Helps operators act before foaming events occur. 💡 Ongoing research links these dynamics to predator–prey cycles, possibly enabling biological control instead of chemical treatments.

🧠 Final Takeaway

Despite many sources of bias and uncertainty:

16S rRNA gene sequencing works beautifully when paired with good sampling, thoughtful design, and validation.

It provides insight into “who is there” — the microbial ecology — forming the foundation for functional understanding, prediction, and control in environmental biotechnology. 🌿🧬💧