clippda

clippda estimates sample sizes for clinical proteomic profiling studies to support detection of biomarkers and optimize study design in cancer observational case-control studies with technical replicates.

Key Features:

• Sample Size Calculation: Computes sample size requirements while accounting for expected heterogeneities, imbalances in subject numbers across phenotypic groups, and within-sample replicate correlations to achieve target statistical power.
• Covariate Adjustment: Incorporates covariate information by modeling the joint distribution of protein expression values and covariates and permitting adjustments based on discretized covariate effects.
• Design Optimization: Evaluates alternative experimental designs and recommends configurations that can reduce required sample sizes, for example favoring balanced studies or specific chip types such as IMAC30 under certain conditions.
• Technical Replicate Handling: Explicitly models technical replicates and non-randomized subject samples in the design and sample size calculations.
• Profiling Technology Consideration: Targets clinical proteomic platforms including MALDI and SELDI profiling when assessing design and sample size implications.

Scientific Applications:

• Cancer biomarker discovery: Supports planning of proteomic studies aimed at identifying biomarkers for potentially curable early-stage cancers.
• Observational case-control studies: Guides sample size and design decisions for human-subject case-control proteomic profiling with technical replicates and covariate data.
• Resource optimization in experimental design: Informs choices that minimize resource use while maintaining power, such as balancing group sizes or selecting chip technologies.

Methodology:

Integrates expression data from multiple clinically defined groups, models the joint distribution of protein expressions and covariates, adjusts for heterogeneities, group-size imbalances, and within-sample replicate correlations, and evaluates alternative study designs to produce sample size recommendations.