Biological vs. Technical Replicate Declaration

The distinction between biological and technical replicates is fundamental to interpreting experimental results. A biological replicate tests whether an effect is present across independent biological units. A technical replicate tests the precision of measurement. Confusing the two produces overstated n and understated uncertainty.

STEP 1 —The Pitfall

Technical replicates —running the same sample on a plate three times —measure the precision of the assay, not the generalizability of the biological effect. They cannot be treated as independent data points in a statistical comparison because they share the same biological source. All three measurements from the same sample have the same underlying biology; their variation is measurement noise, not biological variation.

The misclassification is especially common in cell-based assays (three wells from the same cell passage), Western blots (three lanes from the same lysate), and PCR (three technical duplicates from the same RNA). In each case, the "n" on the figure may represent technical replicates while the figure legend implies biological replicates.

[Image placeholder: Table showing two datasets with identical structure —9 rows, 2 groups. Dataset A: 3 biological replicates x3 technical replicates per group (biological n=3). Dataset B: 9 fully independent biological samples (biological n=9). The same t-test on both gives different SEs —the first is artificially small because technical variance is much lower than biological variance.]

Same table structure, different biological meaning. The replicate type determines what the SE and p-value actually measure.

STEP 2 —Journal Requirement

ARRIVE 2.0: "Clearly state what constitutes an independent replicate in the context of the experiment." Nature Methods editorial: "We require that the type of replicate be stated in the Methods, and that n represents the number of biologically independent samples unless otherwise stated." Cell and other journals now require replicate type disclosure as a condition of publication.

STEP 3 — What Licklider Currently Provides

What the current product does

The current product handles biological vs. technical replicate concerns primarily through the pseudoreplication risk path. Card 17 flags possible pseudoreplication when observation structure suggests non-independent samples, and the user is asked to confirm the study design. The sampling_independence_status and related confirmation state can affect export-facing behavior.

What the product does not do

The product does not provide a dedicated replicate type declaration field (biological / technical / mixed) as a structured input.
The product does not automatically cross-validate a declared replicate type against subject ID structure.
The product does not compute or display biological n vs. technical n separately.
The product does not auto-generate replicate type text for figure legends.

Known limitations

Replicate type is handled implicitly through pseudoreplication risk signals rather than through a dedicated declaration workflow. The researcher is responsible for understanding and correctly reporting the replicate type in their methods text.

[Image placeholder — not current product UI. Illustrative concept of a replicate declaration panel.]

Illustrative concept only. The current product handles replicate concerns via the pseudoreplication confirmation flow (Card 17), not via a dedicated replicate declaration panel.

STEP 4 — Draft Output (Draft / Needs review)

The following is an example of how a researcher might write a replicate type disclosure, not output that Licklider generates automatically.

All experiments were performed in biological triplicate (n = 3
independent experiments, each prepared from separate cell passages
on different days). Within each experiment, measurements were made
in technical duplicate; technical duplicates were averaged within
each experiment prior to statistical analysis. The n shown in figures
represents the number of biologically independent experiments.