#001 — One-Sample Student's t-Test
Statistical Definition
Evaluates whether the mean of a single continuous variable significantly differs from a known or hypothesized population constant under normality.
Bioinformatics Use Case
Testing if normalized log2 fold-change of a housekeeping gene across tumor samples deviates from zero.
Input Variables
1 continuous numeric vector (x); 1 scalar hypothesized mean (mu).
Application
When you have one experimental cohort and need to prove your biomarker's baseline activity differs from a standard reference value, this test delivers a clear, statistically rigorous benchmark.
#002 — Two-Sample Independent Student's t-Test
Statistical Definition
Compares the means of two independent, normally distributed groups assuming equal population variances (homoscedasticity).
Bioinformatics Use Case
Comparing mean normalized expression of a candidate target gene between wild-type and knockout cell lines.
Input Variables
2 independent continuous numeric vectors (x1, x2) with equal variance.
Application
Want to prove your drug treatment shifts target expression compared to untreated controls? Assuming equal variance, this classic test quantifies the exact signal-to-noise ratio between cohorts.
#003 — Welch's Two-Sample t-Test
Statistical Definition
Compares means between two independent groups without assuming equal variances, adjusting degrees of freedom via the Welch-Satterthwaite equation.
Bioinformatics Use Case
Comparing noisy clinical biomarker levels between healthy donors (tight variance) and heterogeneous cancer patients (wide variance).
Input Variables
2 independent continuous numeric vectors (x1, x2) with unequal variances.
Application
Clinical patient cohorts rarely share identical variance with healthy controls. Welch's t-test protects your discovery pipeline from false positives caused by biological heterogeneity.
#004 — Paired Sample t-Test
Statistical Definition
Tests whether the mean difference between paired or matched observations is significantly different from zero.
Bioinformatics Use Case
Evaluating pre-treatment versus post-treatment tumor biopsy gene expression within the exact same patients.
Input Variables
2 matched continuous numeric vectors (x_pre, x_post) of equal length.
Application
By pairing pre- and post-treatment biopsies from the same patients, we eliminate inter-individual baseline noise and isolate the true therapeutic effect with maximum statistical power.
#005 — One-Way Analysis of Variance (ANOVA)
Statistical Definition
Tests whether three or more independent groups have identical population means by partitioning total variance into between-group and within-group components.
Bioinformatics Use Case
Comparing protein abundance across four distinct molecular subtypes of breast cancer (Luminal A, Luminal B, HER2+, Basal).
Input Variables
1 continuous response vector (y); 1 categorical grouping factor (k >= 3 levels).
Application
When screening biomarkers across multiple disease subtypes or drug doses simultaneously, One-Way ANOVA prevents the false-discovery explosion of running multiple pairwise tests.
#006 — Welch's One-Way ANOVA
Statistical Definition
Extension of Welch's t-test to three or more groups, comparing group means without requiring homogeneity of variance across cohorts.
Bioinformatics Use Case
Comparing serum cytokine concentrations across healthy, mild, and severe sepsis cohorts where variance scales with disease severity.
Input Variables
1 continuous response vector (y); 1 categorical factor (k >= 3 levels) with heteroscedastic variances.
Application
Severe disease states introduce massive biological spread. Welch's ANOVA detects true multi-group mean shifts even when variance explodes across clinical stages.
#007 — Two-Way Factorial ANOVA
Statistical Definition
Evaluates the main effects of two categorical factors and their statistical interaction on a continuous outcome.
Bioinformatics Use Case
Testing how Genotype (WT vs. Mutant) and Drug Treatment (Vehicle vs. Inhibitor) jointly affect cell viability, isolating synthetic lethality.
Input Variables
1 continuous response (y); 2 categorical factors (A, B) + interaction term (A:B).
Application
Does your drug only work in patients carrying a specific mutation? Two-Way ANOVA mathematically isolates that exact genotype-by-drug synergy
#008 — Repeated Measures ANOVA
Statistical Definition
Analyzes longitudinal continuous measurements taken on the same subjects across three or more timepoints or conditions.
Bioinformatics Use Case
Tracking longitudinal plasma biomarker decay at 0h, 6h, 24h, and 72h post-dosing in the same animal cohort.
Input Variables
1 continuous outcome (y); 1 within-subject time/condition factor; 1 subject identifier.
Application
Longitudinal trials are expensive; Repeated Measures ANOVA extracts maximum statistical power from small patient or animal cohorts by tracking within-subject trajectories over time.
#009 — Analysis of Covariance (ANCOVA)
Statistical Definition
Compares group means of a continuous outcome while statistically adjusting for the linear effect of one or more continuous confounding covariates.
Bioinformatics Use Case
Comparing tumor volume reduction between two therapies while adjusting for baseline pre-treatment tumor size and patient age.
Input Variables
1 continuous response (y); 1 categorical group factor; 1+ continuous covariates (x).
Application
Baseline differences in tumor size or patient age can obscure a winning drug. ANCOVA mathematically levels the playing field to reveal true treatment efficacy.
#010 — Multivariate Analysis of Variance (MANOVA)
Statistical Definition
Simultaneously tests differences in mean vectors across multiple correlated continuous dependent variables across two or more groups.
Bioinformatics Use Case
Testing whether a metabolic perturbation alters a coordinated panel of 5 tricarboxylic acid (TCA) cycle intermediates simultaneously.
Input Variables
Matrix of p >= 2 correlated continuous outcomes (Y); 1 categorical grouping factor.
Application
Biological pathways move as coordinated networks, not isolated variables. MANOVA captures multi-analyte shifts that individual univariate tests miss ent