Paper - Review

DOI: 10.1371/journal.pcbi.1006102

Abstract

High-throughput data generation platforms
← e.g. 1⃣ mass-spectrometry 2⃣ microarrays 3⃣ second-generation sequencing
→ are susceptible → to batch effects
∵ Run-to-run variation
← in 1⃣ reagents 2⃣ equipment 3⃣ protocols 4⃣ personnel

Batch correction methods
→ are NOT ❌ commonly applied → to microbiome sequencing datasets

❗ : Compare different batch-correction methods
→ applied to microbiome case-control studies

Introduce → a model-free normalization procedure
← where features in case samples → are converted → to percentiles of the equivalent features
← in control samples → prior to pooling data ← across studies

❓: How this percentile-normalization methods
→ compares → to traditional meta-analysis methods → for combining independent p-values
→ to 1⃣ limma 2⃣ ComBat ← widely used batch-correction models
← developed for RNA microarray data

∴ Percentile-normalization
→ is a 1⃣ simple 2⃣ non-parametric approach
→ for 1⃣ correcting batch effects 2⃣ improving sensitivity ← in case-control meta-analyses

---