Skip to content

Sample Preparation and Quality Control

Note: This page is educational. Real laboratory work requires SOPs, biosafety procedures, institutional approvals, and trained supervision.

TL;DR

Sample preparation is the first analysis step. Before PCR, sequencing, proteomics, flow cytometry, or imaging, the biological material has already been filtered by collection, ischemia time, fixation, storage, extraction, dissociation, and QC. Many "computational surprises" are actually sample-prep artifacts.


1. What can happen to a cancer sample

Sample stateCommon downstream useKey risk
Fresh tissueorganoids, flow, single-cell, viable assaysischemia, dissociation bias
Fresh frozenDNA/RNA/protein, spatial, metabolomicscold-chain failure, freeze-thaw
FFPEpathology, IHC, FISH, targeted DNA/RNAfragmentation, formalin artifacts
Blood plasmactDNA, proteins, metaboliteshemolysis, delayed processing
Blood cellsgermline DNA, immune profilinganticoagulant and storage effects
Bone marrowhematologic flow/NGShemodilution, clotting
Effusion/ascitescytology, organoids, cfDNAlow tumor fraction, inflammation

2. Pre-analytical variables

The same tumor can produce different data depending on:

  • time from excision to preservation
  • temperature during transport
  • fixative type and fixation duration
  • tumor cellularity and necrosis
  • stromal and immune admixture
  • blood contamination
  • extraction method
  • freeze-thaw cycles
  • batch order
  • operator and instrument differences

When data look strange, check the chain of custody before blaming biology.


3. QC by assay family

Assay familyQC signals
DNA sequencingDNA amount, fragment size, tumor purity, library yield, depth
RNA-seq / RT-qPCRRNA integrity, DV200/RIN, ribosomal content, mapping rate
Proteomicsprotein yield, digestion efficiency, peptide IDs, missingness
Flow/FACSviability, singlets, compensation, unstained/FMO controls
IHC/IF/FISHtissue preservation, controls, staining batch, scanner QC
Organoids/screensviability, mycoplasma, passage number, growth rate

4. Tumor purity matters

A bulk tumor sample is a mixture:

  • malignant cells
  • fibroblasts
  • immune cells
  • endothelial cells
  • necrosis
  • normal adjacent tissue
  • blood

Low tumor purity can hide mutations, dilute RNA expression, blur methylation signals, and confuse proteomics. High immune infiltration can be biologically meaningful or a confounder depending on the question.


5. Metadata developers should demand

For every sample, try to capture:

MetadataWhy it matters
sample type and anatomic sitebiological comparability
collection time and preservation timedegradation and ischemia
preservation methodFFPE vs frozen vs fresh
tumor percentagevariant detection and expression interpretation
necrosis percentagefailed extraction and artifacts
prior therapytreatment-induced changes
extraction kit/protocolbatch effects
QC metricsreproducibility and filtering
batch and operatorhidden confounding

6. What technologists can build

  • Sample lineage graphs from collection to file.
  • QC dashboards that combine wet-lab metrics with computational metrics.
  • Batch-effect reports before biological interpretation.
  • Metadata validators that block analysis when essential sample fields are missing.
  • Tumor-purity-aware pipelines for mutation, expression, and methylation analysis.

See also

Early public release. Content evolves through continuous review. Questions: [email protected] · CC BY 4.0 where applicable.