CLI tools for pharmaceutical, microbiology and molecular-biology laboratory data — the calculations a QC lab does every day, packaged as a single tested Python library with one entry point.
pip install .
lab-tools --help
lab-tools cfu --help| Domain | Tool | What it does |
|---|---|---|
| Microbiology / QC | cfu |
CFU enumeration from dilution plates (USP ⟨61⟩/⟨62⟩, EP 2.6.12) |
mpn |
Most probable number | |
media-fill |
Aseptic process simulation (USP ⟨1116⟩) | |
growth-curve |
Logistic / Gompertz growth kinetics | |
bioburden-spc |
I-MR control charts (Western Electric) | |
| Sterility / thermal | d-z-f0 |
D-value, Z-value, F0 lethality |
sterility |
Sterility-test sample sizes (USP ⟨71⟩) | |
endotoxin |
Endotoxin limits + MVD (USP ⟨85⟩) | |
| Molecular biology | qpcr |
ΔΔCt relative gene expression |
primers |
Annealing temperature + dimer analysis | |
restriction |
Restriction digest simulation | |
phylogeny |
Neighbor-joining trees | |
| Analytical | elisa |
4PL standard curves |
lod-loq |
LOD / LOQ (ICH Q2(R1)) | |
hplc-sst |
HPLC system suitability | |
| Lab operations | hemocytometer |
Cell count / viability |
pipette-cal |
Pipette calibration (ISO 8655) |
A 1:10 serial dilution plated at 0.1 mL per plate with duplicate counts:
| Dilution factor | Replicate A | Replicate B |
|---|---|---|
| 0.01 (10⁻²) | 245 | 231 |
| 0.001 (10⁻³) | 42 | 38 |
All four counts are inside the USP/EP countable range of 30–300 CFU. For each plate:
CFU/mL = colony count / (volume plated in mL × dilution factor)
245 / (0.1 × 0.01) = 245,000 CFU/mL
231 / (0.1 × 0.01) = 231,000 CFU/mL
42 / (0.1 × 0.001) = 420,000 CFU/mL
38 / (0.1 × 0.001) = 380,000 CFU/mL
The reported mean across the four usable plates is:
(245,000 + 231,000 + 420,000 + 380,000) / 4 = 319,000 CFU/mL
The between-plate CV is about 29.82% (sample standard deviation ≈ 95,118 CFU/mL).
Run the same calculation with the one-line command:
lab-tools cfu quick --counts 245,231,42,38 --dilutions 0.01,0.001 --replicates 2 --volume 0.1The report prints CFU/mL: 3.19e+05 and the same uncertainty and dilution details.
pip install -e . pytest
python -m pytest -qThe project grows through a daily backlog-driven automation (see
BACKLOG.md and scripts/): one item is implemented and committed per day,
only if the test suite stays green.
MIT © Collins Amatu Gorgerat