PIIKA 2.5

PIIKA 2.5 performs analysis and quality control of kinome peptide microarray data to detect and compare phosphorylation-mediated kinase activity across samples.

Key Features:

• Spot Size Alert: A metric that flags improper spot sizes on peptide microarrays that can affect raw intensity measurements.
• Inter-array Comparison: Identification of outlier arrays by comparing arrays across experiments to detect technical artifacts.
• Background Scaling: A background scaling method that reduces spatial biases within single arrays compared to traditional background subtraction.
• Cluster Evaluation: Statistical evaluation of how well groups of samples cluster together.
• Peptide Pattern Identification: Identification of sets of peptides with consistent phosphorylation patterns across sample groups.
• Hierarchical Clustering with Bootstrapping: Hierarchical clustering analysis supplemented with bootstrapping for statistical validation.
• Pairwise t-test Diagnostics: Pairwise t-tests with estimation of false negative probabilities and calculation of positive and negative predictive values.
• Reproducibility Assessment: Evaluation metrics for experimental reproducibility across arrays and experiments.
• Volcano and Scatter Plots: Generation of volcano plots and scatterplots for differential signal visualization.
• Three-dimensional Principal Component Analysis: Three-dimensional principal component analysis for dimensionality reduction and visualization.

Scientific Applications:

• Kinome Peptide Array Analysis: Extraction of biological information from kinome peptide array data for high-throughput study of kinase activity.
• Cellular Kinase Activity Profiling: Comparison of phosphorylation-mediated kinase activity across samples and experimental conditions.
• Batch-effect Detection and Correction: Detection and reduction of technical and spatial biases to improve comparability across experiments.
• Signaling Pathway Studies: Analysis of peptide-level phosphorylation patterns to support studies of phosphorylation-mediated cellular signaling pathways.
• Treatment Group Clustering and Reproducibility: Improvement of clustering of treatment groups and enhancement of reproducibility across experiments.

Methodology:

Methods explicitly include spot-size metrics, inter-array comparison, background scaling versus traditional background subtraction, hierarchical clustering with bootstrapping, pairwise t-tests with estimation of false negative probabilities and positive/negative predictive values, identification of peptides with consistent phosphorylation patterns, volcano plots, scatterplots, and three-dimensional principal component analysis.