Wet Lab 101: From Test Tube to FASTQ

Under scientific review

This introductory page is under scientific review. Use it for orientation, not clinical decision-making.

Before bioinformatics algorithms can align reads or call variants, biological tissue must undergo a physical and chemical transformation. For software engineers, the Wet Lab is the physical data ingestion pipeline that ingests biological hardware (cells, tissues, blood) and digitizes it into raw data files (.fastq).

Understanding this physical ETL process is crucial: sequencing artifacts, coverage dropouts, and batch effects almost always originate in the wet lab stage.

---

1. Sample Collection & Preservation

Biological samples are the input payload. How they are preserved dictates data quality:

• FFPE (Formalin-Fixed Paraffin-Embedded) — Lossy / JPEG Compression: Tissue is fixed in formalin and embedded in wax. It can be stored at room temperature for decades, making it the most common clinical archive. However, formalin causes chemical cross-linking and DNA fragmentation (reducing read length and introducing C→T artifact transition mutations).
• Fresh-Frozen (FF) — Lossless / PNG Compression: Tissue is flash-frozen in liquid nitrogen (-196°C). DNA/RNA remains high-molecular-weight and intact, making FF the gold standard for high-depth sequencing and long-read technologies.
• Liquid Biopsy — Streaming Input: Cell-free DNA (cfDNA) shed by tumor cells into blood plasma. Requires specialized tubes (e.g., Streck) to prevent normal white blood cells from lysing and diluting the tumor signal.

---

2. Extraction: Cellular Decompression

Extracting DNA/RNA requires stripping away cellular membranes, proteins, and lipids:

1. Cell Lysis (unzip): Detergents and chaotropic salts break cell and nuclear membranes.
2. Protein Digestion (garbage collection): Proteinase K enzymes digest histones and cellular proteins bound to nucleic acids.
3. Phase Separation / Column Capture (grep): Silica-membrane spin columns bind DNA/RNA under high salt conditions while contaminants are washed away.
4. Elution (write to disk): Pure nucleic acids are eluted in nuclease-free water or buffer.

---

3. Quality Control (QC): Unit Testing the Input

Before spending thousands of dollars on sequencing, nucleic acids must pass quality gates:

• NanoDrop (Spectrophotometry) — Header Check: Measures absorbance ratios (260/280nm for protein contamination, 260/230nm for salt/phenol contamination). Quick but overestimates concentration.
• Qubit (Fluorometry) — Exact Checksum: Uses dye molecules that fluoresce only when bound to intact double-stranded DNA or RNA. Provides exact concentration (ng/µL).
• Bioanalyzer / TapeStation — Payload Integrity Check: Electrophoresis separates molecules by size. Yields a DIN (DNA Integrity Number) or RIN (RNA Integrity Number) from 1 to 10. Samples with RIN < 7 are degraded and require specialized library preparation.

---

4. Library Preparation: Formatting for the Sequencer

Raw DNA cannot be loaded directly onto a sequencer. Library Prep formats the raw DNA into machine-readable constructs:

[ P5 Adapter ] -- [ Index 1 / Barcode ] -- [ Insert (Target DNA) ] -- [ Index 2 ] -- [ P7 Adapter ]

1. Fragmentation: DNA is sheared via ultrasonication (Covaris) or enzymatic cleavage into uniform chunks (~300-500 bp for short-read sequencing).
2. End Repair & A-Tailing: Sheared ends are blunted and an 'A' base is added to the 3' ends.
3. Adapter Ligation (HTTP Headers): Y-shaped synthetic DNA adapters are ligated. These contain:
   • Flow Cell Binding Sites (P5/P7): Anchors that bind to the sequencer hardware.
   • Sample Barcodes / Indexes (i5/i7): Unique 8-10 bp identifiers allowing 96+ patient samples to be pooled into a single run (Multiplexing / Multi-tenancy).
   • Sequencing Primer Binding Sites: Launch pads for the sequencing reaction.
4. PCR Amplification (cp -r): Amplifies the library. Excessive cycles introduce PCR duplicates (duplicate reads that must be removed downstream during alignment).

---

5. Flow Cell Loading & Demultiplexing

The library pool is loaded onto a Flow Cell (a glass slide microfluidic chip coated with billions of complementary oligonucleotides):

1. Bridge Amplification / Exclusion Amplification: Single molecules bind to the flow cell surface and clonally amplify into physical clusters (~1,000 identical copies per spot).
2. Sequencing Run: The instrument cycles through fluorescent nucleotides, capturing high-resolution images at each base step (Sequencing-by-Synthesis).
3. Demultiplexing (bcl2fastq / DRAGEN): Raw optical image data (.bcl binary files) are converted into base calls and split by sample barcode into individual .fastq.gz files.

---

Summary Workflow

$$\text{Biopsy} \xrightarrow{\text{Lysis}} \text{Extraction} \xrightarrow{\text{Qubit/RIN}} \text{Shearing} \xrightarrow{\text{Adapter Ligation}} \text{Flow Cell} \xrightarrow{\text{Demux}} \text{FASTQ}$$

---

References

1. Head, S. R., et al. (2014). Library construction for next-generation sequencing: overviews and challenges. Biotechniques, 56(2), 61-77.
2. Illumina Inc. (2020). An Introduction to Next-Generation Sequencing Technology. San Diego, CA.
3. Sambrook, J., & Russell, D. W. (2001). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press.