Power analysis

Version Compatibility

Reference examples tested with: RNASeqPower 1.42+, pwr 1.3+

Before using code patterns, verify installed versions match. If versions differ:

• R: packageVersion("<pkg>") then ?function_name to verify parameters

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Power Analysis for Sequencing Experiments

"How many replicates do I need for RNA-seq?" → Calculate statistical power or minimum sample size given sequencing depth, biological variability, and expected effect size.

• R: RNASeqPower::rnapower(), pwr::pwr.t.test()

Core Concept

Power = probability of detecting a true effect. Underpowered studies waste resources; overpowered studies are inefficient.

RNA-seq Power Analysis

Goal: Determine whether a planned RNA-seq experiment has sufficient statistical power to detect biologically meaningful fold changes, or calculate the minimum sample size needed for a target power.

Approach: Provide sequencing depth, biological coefficient of variation, expected fold change, and significance level to rnapower, which uses a negative binomial model to compute power or required sample size.

library(RNASeqPower)

# Typical parameters
# - depth: sequencing depth per sample (reads/gene)
# - cv: biological coefficient of variation (0.1-0.4 typical)
# - effect: fold change to detect (1.5 = 50% change)
# - alpha: significance level (0.05 standard)

# Calculate power for given sample size
rnapower(depth = 20, n = 3, cv = 0.4, effect = 2, alpha = 0.05)

# Calculate required samples for target power
rnapower(depth = 20, cv = 0.4, effect = 2, alpha = 0.05, power = 0.8)

CV Guidelines

Experiment Type	Typical CV	Notes
Cell lines	0.1-0.2	Low variability
Inbred mice	0.2-0.3	Moderate
Human samples	0.3-0.5	High variability
Primary cells	0.3-0.4	Donor-dependent

ATAC-seq Power (ssizeRNA)

library(ssizeRNA)

# For differential accessibility
size.zhao(m = 10000, m1 = 500, fc = 2, fdr = 0.05, power = 0.8,
          mu = 10, disp = 0.1)

Quick Reference

Effect Size	Recommended n (CV=0.4)
4-fold	3 per group
2-fold	5-6 per group
1.5-fold	10-12 per group
1.25-fold	20+ per group

Related Skills

• experimental-design/sample-size - Detailed sample size calculations
• experimental-design/batch-design - Accounting for batch effects in design
• differential-expression/deseq2-basics - Running the actual DE analysis