SpaTemHTP

SpaTemHTP processes time-series data from high-throughput phenotyping (HTP) platforms to produce spatially adjusted genotypic estimates, detect growth-phase change points, and derive smooth genotype growth curves for crop trait analysis.

Key Features:

• Implementation: Implemented as an R package for processing and analysis of HTP time-series data.
• Data Processing Modules: Sequential pipeline modules include Outlier Detection, Missing Value Imputation capable of handling up to 50% missing values, and Mixed-Model Genotype Adjusted Means computation with spatial adjustments.
• Robustness: Pipeline is reported to tolerate contamination rates of approximately 20–30% without compromising analysis integrity.
• Change-Point Analysis: Identifies growth phases in genotype time series to optimize timing for detecting significant genotypic differences.
• Clustering and Statistical Analysis: Clusters genotypes using estimated genotypic values during optimal growth phases and validates cluster consistency with two-way ANOVA.

Scientific Applications:

• Outdoor HTP time-series analysis: Extracts smooth genotype growth curves and refines noisy and incomplete datasets typical of outdoor phenotyping platforms.
• Trait- and crop-specific evaluation: Applied to 3D leaf area, projected leaf area, and plant height in chickpea and sorghum across two seasons to support trait analysis and breeding decisions.

Methodology:

Sequential application of outlier detection, missing-value imputation, spatially adjusted mixed-model genotype mean estimation, change-point analysis, clustering, and two-way ANOVA, evaluated on real-data analyses and simulations for 3D leaf area, projected leaf area, and plant height in chickpea and sorghum over two seasons.