All examples · Example 32 of 34

PCAOmicsStatistical report

PCA: Sample Clustering

Principal component analysis of 30 samples (6 conditions × 5 replicates) on 50 features; points coloured by condition.

Key findings

PC1 and PC2 explain 24.9% and 17.1% of the variance.
48 of 50 features differ between conditions after Benjamini–Hochberg correction (43 after Bonferroni); the strongest is Feature39 (ω² = 0.892).

75 LabTools stats calls

7 of 15 stats methods

LabTools 0.14.3

600 dpi PNG Data (CSV) Raw LabTools output

Which of the 50 features separate the conditions?

PCA shows that the 6 conditions separate; this asks which features drive it. Each feature gets a one-way ANOVA across the conditions (n = 5 samples each), and the 50 p-values are corrected as one family.

50 tests, α = 0.05

Correction	Controls	Tests	Raw p < 0.05	Significant after correction
Bonferroni	family-wise error	50	48	43
Holm	family-wise error	50	48	47
Benjamini–Hochberg	false discovery rate	50	48	48
Benjamini–Yekutieli	false discovery rate	50	48	47

LabTools calls (5)

stats.anova

labtools stats anova --groups @Feature01.json --no-assumption-checks

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method bonferroni --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method holm --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method benjamini-hochberg --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method benjamini-yekutieli --alpha 0.05

The same method is available as an MCP tool (stats_anova) and from the desktop app's statistics panel.

Top 15 by raw p

Feature	F	ω²	Raw p	Bonferroni	Holm	BH	BY
Feature39	50.45	0.892	5.9 × 10⁻¹²	3.0 × 10⁻¹⁰	3.0 × 10⁻¹⁰	3.0 × 10⁻¹⁰	1.3 × 10⁻⁹
Feature11	44.74	0.879	2.2 × 10⁻¹¹	1.1 × 10⁻⁹	1.1 × 10⁻⁹	5.3 × 10⁻¹⁰	2.4 × 10⁻⁹
Feature16	43.13	0.875	3.2 × 10⁻¹¹	1.6 × 10⁻⁹	1.5 × 10⁻⁹	5.3 × 10⁻¹⁰	2.4 × 10⁻⁹
Feature31	40.29	0.868	6.6 × 10⁻¹¹	3.3 × 10⁻⁹	3.1 × 10⁻⁹	8.2 × 10⁻¹⁰	3.7 × 10⁻⁹
Feature25	36.52	0.855	1.8 × 10⁻¹⁰	9.2 × 10⁻⁹	8.5 × 10⁻⁹	1.8 × 10⁻⁹	8.0 × 10⁻⁹
Feature32	36.02	0.854	2.1 × 10⁻¹⁰	1.1 × 10⁻⁸	9.6 × 10⁻⁹	1.8 × 10⁻⁹	8.0 × 10⁻⁹
Feature18	34.3	0.847	3.5 × 10⁻¹⁰	1.8 × 10⁻⁸	1.6 × 10⁻⁸	2.5 × 10⁻⁹	1.1 × 10⁻⁸
Feature10	32.98	0.842	5.3 × 10⁻¹⁰	2.7 × 10⁻⁸	2.3 × 10⁻⁸	3.3 × 10⁻⁹	1.5 × 10⁻⁸
Feature15	32.02	0.838	7.2 × 10⁻¹⁰	3.6 × 10⁻⁸	3.0 × 10⁻⁸	4.0 × 10⁻⁹	1.8 × 10⁻⁸
Feature40	31.67	0.836	8.1 × 10⁻¹⁰	4.0 × 10⁻⁸	3.3 × 10⁻⁸	4.0 × 10⁻⁹	1.8 × 10⁻⁸
Feature41	31.21	0.834	9.4 × 10⁻¹⁰	4.7 × 10⁻⁸	3.8 × 10⁻⁸	4.3 × 10⁻⁹	1.9 × 10⁻⁸
Feature13	27.45	0.815	3.4 × 10⁻⁹	1.7 × 10⁻⁷	1.3 × 10⁻⁷	1.4 × 10⁻⁸	6.4 × 10⁻⁸
Feature34	26.09	0.807	5.7 × 10⁻⁹	2.8 × 10⁻⁷	2.2 × 10⁻⁷	2.2 × 10⁻⁸	9.9 × 10⁻⁸
Feature44	25.54	0.804	7.0 × 10⁻⁹	3.5 × 10⁻⁷	2.6 × 10⁻⁷	2.5 × 10⁻⁸	1.1 × 10⁻⁷
Feature45	21.63	0.775	3.5 × 10⁻⁸	1.8 × 10⁻⁶	1.3 × 10⁻⁶	1.2 × 10⁻⁷	5.3 × 10⁻⁷

LabTools notes

5 comparison(s) were significant on the raw p-values but are not after correction. Report the adjusted values.
Holm controls the same familywise error rate as Bonferroni and rejects at least as much, so there is no statistical reason to prefer bonferroni.
1 comparison(s) were significant on the raw p-values but are not after correction. Report the adjusted values.

Top feature: Feature39

Location, spread and shape for every group — the first thing to read before choosing a test. Skewness near 0 and excess kurtosis near 0 are what a normal distribution looks like.

Group	n	Mean	SD	SEM	Median	Q1 – Q3	Min – Max	Skewness	Excess kurtosis
Control	5	1.878	0.6674	0.2985	1.702	1.35 – 2.327	1.22 – 2.793	0.584	-1.65
Drug A	5	-1.656	1.02	0.4562	-1.334	-1.668 – -1.27	-3.359 – -0.651	-1.5	2.91
Drug B	5	0.208	0.9924	0.4438	0.095	-0.24 – 0.378	-0.942 – 1.749	0.875	1.57
Drug A+B	5	-4.927	0.7578	0.3389	-5.009	-5.075 – -4.463	-6.05 – -4.037	-0.59	0.585
Stress	5	2.747	1.165	0.5209	2.453	1.995 – 2.983	1.671 – 4.634	1.33	1.74
Recovery	5	-2.626	0.7574	0.3387	-2.538	-2.897 – -2.056	-3.771 – -1.866	-0.862	0.223

LabTools call

stats.describe

labtools stats describe --groups @groups.json

The same method is available as an MCP tool (stats_describe) and from the desktop app's statistics panel.

Feature39: which conditions differ?

One-way ANOVA compares means and assumes normal residuals with equal variances; Kruskal–Wallis compares ranks and assumes neither. Both only say the groups differ somewhere — the pairwise table below says where.

Test	Statistic	df	p	Effect size
One-way ANOVA	F = 50.45	5, 24	5.9 × 10⁻¹²	ω² = 0.892 · η² = 0.913 · Cohen's f = 3.24
Kruskal–Wallis H	H = 26.4	5	7.5 × 10⁻⁵	ε² = 0.91

LabTools calls (2)

stats.anova

labtools stats anova --groups @groups.json

stats.kruskal

labtools stats kruskal --groups @groups.json

The same method is available as an MCP tool (stats_anova) and from the desktop app's statistics panel.

LabTools notes

Group 'Control' has n = 5; normality screening has very low power at this size, so a non-significant Shapiro–Wilk is weak evidence of normality.

Which groups differ? (pairwise, corrected)

5 comparisons (each group against Control), each run as a Welch t-test and a Mann–Whitney test. Testing many pairs inflates false positives, so every p-value is corrected with Holm (family-wise) and Benjamini–Hochberg (false-discovery rate) through stats.adjust.

Comparison	Mean diff.	95% CI	Hedges' g	Welch p	Holm	BH	Mann–Whitney p	Holm	Cliff's δ
Control vs Drug A	3.535	[2.242, 4.828]	3.7	0.0004	0.001	0.0006	0.012	0.061	1
Control vs Drug B	1.67	[0.4058, 2.935]	1.78	0.017	0.034	0.021	0.060	0.12	0.76
Control vs Drug A+B	6.805	[5.761, 7.849]	8.61	4.4 × 10⁻⁷	2.2 × 10⁻⁶	2.2 × 10⁻⁶	0.012	0.061	1
Control vs Stress	-0.8688	[-2.317, 0.5796]	-0.827	0.20	0.20	0.20	0.21	0.21	-0.52
Control vs Recovery	4.504	[3.46, 5.548]	5.7	9.7 × 10⁻⁶	3.9 × 10⁻⁵	2.4 × 10⁻⁵	0.012	0.061	1

LabTools calls (3)

stats.ttest

labtools stats ttest --a @Control.json --b @Drug_A.json --test welch --no-assumption-checks

stats.mannwhitney

labtools stats mannwhitney --a @Control.json --b @Drug_A.json

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method holm

The same method is available as an MCP tool (stats_ttest) and from the desktop app's statistics panel.

Design sensitivity

d = 2.02

Minimum detectable effect (80% power, α = 0.05)

n = 5 vs 5

10.8%

Power to detect d = 0.5

20.1%

Power to detect d = 0.8

38.6%

Power to detect d = 1.2

LabTools calls (2)

stats.power

labtools stats power --n1 5 --n2 5 --mode mde --power 0.8

stats.power

labtools stats power --n1 5 --n2 5 --effect-size 0.5

The same method is available as an MCP tool (stats_power) and from the desktop app's statistics panel.

Methods paragraph

Written to go straight into a manuscript.

PC1 and PC2 explained 24.9% and 17.1% of the variance; replicates of each condition clustered together.

Every number on this page was computed by the LabTools statistics engine (version 0.14.3) from the example's CSV on 2026-09-23. Where a value was derived before the data reached LabTools (replicate means, paired differences, log odds ratios), the text says so. The data are simulated from a fixed random seed.

Get started

Run the same analysis on your data

Every test on this page is built into LabTools — from the statistics panel, the command line, or an AI assistant through MCP.

Download free More examples

LabToolsThe Research Infrastructure

Download

All examples · Example 32 of 34

PCAOmicsStatistical report

PCA: Sample Clustering

Principal component analysis of 30 samples (6 conditions × 5 replicates) on 50 features; points coloured by condition.

Key findings

PC1 and PC2 explain 24.9% and 17.1% of the variance.
48 of 50 features differ between conditions after Benjamini–Hochberg correction (43 after Bonferroni); the strongest is Feature39 (ω² = 0.892).

75 LabTools stats calls

7 of 15 stats methods

LabTools 0.14.3

600 dpi PNG Data (CSV) Raw LabTools output

Which of the 50 features separate the conditions?

50 tests, α = 0.05

Correction	Controls	Tests	Raw p < 0.05	Significant after correction
Bonferroni	family-wise error	50	48	43
Holm	family-wise error	50	48	47
Benjamini–Hochberg	false discovery rate	50	48	48
Benjamini–Yekutieli	false discovery rate	50	48	47

LabTools calls (5)

stats.anova

labtools stats anova --groups @Feature01.json --no-assumption-checks

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method bonferroni --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method holm --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method benjamini-hochberg --alpha 0.05

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method benjamini-yekutieli --alpha 0.05

The same method is available as an MCP tool (stats_anova) and from the desktop app's statistics panel.

Top 15 by raw p

Feature	F	ω²	Raw p	Bonferroni	Holm	BH	BY
Feature39	50.45	0.892	5.9 × 10⁻¹²	3.0 × 10⁻¹⁰	3.0 × 10⁻¹⁰	3.0 × 10⁻¹⁰	1.3 × 10⁻⁹
Feature11	44.74	0.879	2.2 × 10⁻¹¹	1.1 × 10⁻⁹	1.1 × 10⁻⁹	5.3 × 10⁻¹⁰	2.4 × 10⁻⁹
Feature16	43.13	0.875	3.2 × 10⁻¹¹	1.6 × 10⁻⁹	1.5 × 10⁻⁹	5.3 × 10⁻¹⁰	2.4 × 10⁻⁹
Feature31	40.29	0.868	6.6 × 10⁻¹¹	3.3 × 10⁻⁹	3.1 × 10⁻⁹	8.2 × 10⁻¹⁰	3.7 × 10⁻⁹
Feature25	36.52	0.855	1.8 × 10⁻¹⁰	9.2 × 10⁻⁹	8.5 × 10⁻⁹	1.8 × 10⁻⁹	8.0 × 10⁻⁹
Feature32	36.02	0.854	2.1 × 10⁻¹⁰	1.1 × 10⁻⁸	9.6 × 10⁻⁹	1.8 × 10⁻⁹	8.0 × 10⁻⁹
Feature18	34.3	0.847	3.5 × 10⁻¹⁰	1.8 × 10⁻⁸	1.6 × 10⁻⁸	2.5 × 10⁻⁹	1.1 × 10⁻⁸
Feature10	32.98	0.842	5.3 × 10⁻¹⁰	2.7 × 10⁻⁸	2.3 × 10⁻⁸	3.3 × 10⁻⁹	1.5 × 10⁻⁸
Feature15	32.02	0.838	7.2 × 10⁻¹⁰	3.6 × 10⁻⁸	3.0 × 10⁻⁸	4.0 × 10⁻⁹	1.8 × 10⁻⁸
Feature40	31.67	0.836	8.1 × 10⁻¹⁰	4.0 × 10⁻⁸	3.3 × 10⁻⁸	4.0 × 10⁻⁹	1.8 × 10⁻⁸
Feature41	31.21	0.834	9.4 × 10⁻¹⁰	4.7 × 10⁻⁸	3.8 × 10⁻⁸	4.3 × 10⁻⁹	1.9 × 10⁻⁸
Feature13	27.45	0.815	3.4 × 10⁻⁹	1.7 × 10⁻⁷	1.3 × 10⁻⁷	1.4 × 10⁻⁸	6.4 × 10⁻⁸
Feature34	26.09	0.807	5.7 × 10⁻⁹	2.8 × 10⁻⁷	2.2 × 10⁻⁷	2.2 × 10⁻⁸	9.9 × 10⁻⁸
Feature44	25.54	0.804	7.0 × 10⁻⁹	3.5 × 10⁻⁷	2.6 × 10⁻⁷	2.5 × 10⁻⁸	1.1 × 10⁻⁷
Feature45	21.63	0.775	3.5 × 10⁻⁸	1.8 × 10⁻⁶	1.3 × 10⁻⁶	1.2 × 10⁻⁷	5.3 × 10⁻⁷

LabTools notes

5 comparison(s) were significant on the raw p-values but are not after correction. Report the adjusted values.
Holm controls the same familywise error rate as Bonferroni and rejects at least as much, so there is no statistical reason to prefer bonferroni.
1 comparison(s) were significant on the raw p-values but are not after correction. Report the adjusted values.

Top feature: Feature39

Location, spread and shape for every group — the first thing to read before choosing a test. Skewness near 0 and excess kurtosis near 0 are what a normal distribution looks like.

Group	n	Mean	SD	SEM	Median	Q1 – Q3	Min – Max	Skewness	Excess kurtosis
Control	5	1.878	0.6674	0.2985	1.702	1.35 – 2.327	1.22 – 2.793	0.584	-1.65
Drug A	5	-1.656	1.02	0.4562	-1.334	-1.668 – -1.27	-3.359 – -0.651	-1.5	2.91
Drug B	5	0.208	0.9924	0.4438	0.095	-0.24 – 0.378	-0.942 – 1.749	0.875	1.57
Drug A+B	5	-4.927	0.7578	0.3389	-5.009	-5.075 – -4.463	-6.05 – -4.037	-0.59	0.585
Stress	5	2.747	1.165	0.5209	2.453	1.995 – 2.983	1.671 – 4.634	1.33	1.74
Recovery	5	-2.626	0.7574	0.3387	-2.538	-2.897 – -2.056	-3.771 – -1.866	-0.862	0.223

LabTools call

stats.describe

labtools stats describe --groups @groups.json

The same method is available as an MCP tool (stats_describe) and from the desktop app's statistics panel.

Feature39: which conditions differ?

Test	Statistic	df	p	Effect size
One-way ANOVA	F = 50.45	5, 24	5.9 × 10⁻¹²	ω² = 0.892 · η² = 0.913 · Cohen's f = 3.24
Kruskal–Wallis H	H = 26.4	5	7.5 × 10⁻⁵	ε² = 0.91

LabTools calls (2)

stats.anova

labtools stats anova --groups @groups.json

stats.kruskal

labtools stats kruskal --groups @groups.json

The same method is available as an MCP tool (stats_anova) and from the desktop app's statistics panel.

LabTools notes

Group 'Control' has n = 5; normality screening has very low power at this size, so a non-significant Shapiro–Wilk is weak evidence of normality.

Which groups differ? (pairwise, corrected)

Comparison	Mean diff.	95% CI	Hedges' g	Welch p	Holm	BH	Mann–Whitney p	Holm	Cliff's δ
Control vs Drug A	3.535	[2.242, 4.828]	3.7	0.0004	0.001	0.0006	0.012	0.061	1
Control vs Drug B	1.67	[0.4058, 2.935]	1.78	0.017	0.034	0.021	0.060	0.12	0.76
Control vs Drug A+B	6.805	[5.761, 7.849]	8.61	4.4 × 10⁻⁷	2.2 × 10⁻⁶	2.2 × 10⁻⁶	0.012	0.061	1
Control vs Stress	-0.8688	[-2.317, 0.5796]	-0.827	0.20	0.20	0.20	0.21	0.21	-0.52
Control vs Recovery	4.504	[3.46, 5.548]	5.7	9.7 × 10⁻⁶	3.9 × 10⁻⁵	2.4 × 10⁻⁵	0.012	0.061	1

LabTools calls (3)

stats.ttest

labtools stats ttest --a @Control.json --b @Drug_A.json --test welch --no-assumption-checks

stats.mannwhitney

labtools stats mannwhitney --a @Control.json --b @Drug_A.json

stats.adjust

labtools stats adjust --p-values @p_values.json --labels @labels.json --method holm

The same method is available as an MCP tool (stats_ttest) and from the desktop app's statistics panel.

Design sensitivity

d = 2.02

Minimum detectable effect (80% power, α = 0.05)

n = 5 vs 5

10.8%

Power to detect d = 0.5

20.1%

Power to detect d = 0.8

38.6%

Power to detect d = 1.2

LabTools calls (2)

stats.power

labtools stats power --n1 5 --n2 5 --mode mde --power 0.8

stats.power

labtools stats power --n1 5 --n2 5 --effect-size 0.5

The same method is available as an MCP tool (stats_power) and from the desktop app's statistics panel.

Methods paragraph

Written to go straight into a manuscript.

PC1 and PC2 explained 24.9% and 17.1% of the variance; replicates of each condition clustered together.

Get started

Run the same analysis on your data

Every test on this page is built into LabTools — from the statistics panel, the command line, or an AI assistant through MCP.

Download free More examples