Loading Video...
NTHRYS
Arrow

Multiple Testing, FDR Control & Batch Effects Training | Biostatistics & Omics Analytics

NTHRYS >> Services >> Academic Services >> Training Programs >> Bioinformatics Training >> Biostatistics, AI/ML & Reproducible Omics Analytics >> Multiple Testing, FDR Control & Batch Effects Training | Biostatistics & Omics Analytics

Multiple Testing, FDR Control & Batch Effects — Hands-on

Learn how to control false positives in high dimensional omics while handling real world batch effects. This module covers family wise error, FDR procedures, independent filtering and practical batch correction, implemented in R and Python using real gene expression and other omics datasets.

Multiple Testing, FDR Control & Batch Effects
Help Desk · WhatsApp
Session 1
Fee: Rs 8800
Multiple Testing & Family-Wise Error
  • Why multiple testing matters in omics
  • thousands of features false positives vs discovery FWER vs FDR
  • Classical FWER control
  • Bonferroni & Holm family definition conservative vs powerful
  • Permutation & resampling ideas
  • maxT/minP approaches null distributions computational considerations
Session 2
Fee: Rs 11800
FDR Methods in High-Dimensional Omics
  • Benjamini–Hochberg and variants
  • BH & BY procedures q-values & local FDR dependent tests & structure
  • Independent filtering & weighting
  • filtering low-count features p-value weighting trade-off: power vs bias
  • Practical omics workflows
  • gene lists with FDR control volcano & MA plots reporting thresholds & caveats
Session 3
Fee: Rs 14800
Batch Effects: Detection & Correction
  • Sources and signatures of batch effects
  • lab / run / site effects PCA and clustering diagnostics metadata audits
  • Correction strategies
  • ComBat style adjustments including batch in models sva / RUV style methods
  • Common pitfalls & sensitivity checks
  • over-correction risks confounded design issues pre/post batch QC panels
Session 4
Fee: Rs 18800
Integrated Workflow: DE, FDR & Batches
  • End-to-end case study with real data
  • from raw matrix to gene list
  • Reproducible R & Python pipelines
  • R limma / edgeR style GLMs Python pandas & statsmodels scripts & notebooks
  • Deliverables: QC, FDR and batch report
  • pre/post batch plots FDR controlled hit list workflow documentation


PDF