Lab Methods 101
Note: This section explains laboratory methods conceptually. It is not a protocol manual, safety manual, or substitute for local SOPs, biosafety review, clinical validation, or supervision.
TL;DR
Lab methods are the physical and biochemical interfaces between a biological question and a dataset. Omics pages explain data layers; this section explains how those data are produced. The central move is choosing a method that matches the question: DNA sequence, RNA abundance, protein amount, cell phenotype, tissue location, model behavior, or perturbation effect.
Decision tree
| If you want to know... | Start with... | Typical output |
|---|---|---|
| Is the sample usable at all? | Sample preparation and QC | purity, integrity, viability, batch metadata |
| Is this DNA/RNA sequence present? | PCR, qPCR, dPCR | amplicon, Ct/Cq, copies/uL |
| What is the exact sequence? | Sanger, NGS, long-read sequencing | FASTQ, BAM/CRAM, VCF |
| Is this protein present or changed? | Western, ELISA, mass spectrometry | bands, concentrations, peptide IDs |
| Which cells are in the sample? | Flow cytometry, FACS, CyTOF | gated populations, sorted cells, marker matrices |
| Where is a marker inside tissue? | IHC, IF, FISH, RNAscope, spatial | stained slides, images, spatial matrices |
| What happens when biology is perturbed? | Organoids, PDX, GEMM, CRISPR screens | phenotypes, viability, sgRNA counts, model response |
The method stack
| Step | Question |
|---|---|
| 1. Biological question | What do we need to know? |
| 2. Sample and preservation | What material can still answer it? |
| 3. Assay choice | Which method matches the signal? |
| 4. Wet-lab QC | Did the experiment work technically? |
| 5. Raw data | What file or measurement was produced? |
| 6. Computational processing | How is noise, bias, and scale handled? |
| 7. Interpretation | What conclusion is justified? |
The most common failure is not a bad algorithm. It is a mismatch between the biological question, sample quality, assay limits, and interpretation.
Crosswalk to computation
| Lab object | Computational object |
|---|---|
| DNA/RNA extraction | QC metrics, concentration, integrity |
| PCR amplicon | Ct/Cq, melt curve, droplet counts, allele fraction |
| Sequencing library | FASTQ, quality scores, index balance |
| Tissue section | whole-slide image, ROI masks, cell coordinates |
| Flow panel | FCS file, compensation matrix, gating strategy |
| CRISPR screen | sgRNA count matrix, depletion/enrichment scores |
Before choosing a method
Ask these first:
- What sample state is available: fresh, frozen, FFPE, blood, marrow, effusion, organoid, or cell line?
- Is the biological signal DNA, RNA, protein, cell state, tissue location, or function?
- Does the method need viable cells?
- Is the question targeted or discovery-oriented?
- Is spatial context required?
- What minimum QC metric would make the result interpretable?
- What file type will the computational pipeline receive?