iSeqQC

iSeqQC performs expression-based quality control for RNA sequencing data from next-generation sequencing (NGS) to identify biases and outliers that may affect downstream analyses.

Key Features:

• Outlier Detection: Identifies sample outliers arising from batch effects or phenotypic dissimilarities using expression-level metrics.
• Statistical Approaches: Implements unsupervised clustering, agglomerative hierarchical clustering, and correlation analyses using Pearson and Spearman coefficients.
• Summary Statistics: Computes and reports summary statistics for sample-level expression data.
• Counts Distribution Analysis: Evaluates counts distribution across samples to reveal library composition differences.
• Mapped Read Density: Assesses mapped read density as a feature for quality evaluation.
• Housekeeping Gene Expression Analysis: Analyzes housekeeping gene expression to assess sample quality and consistency.
• PCA Variance Analysis: Calculates principal component analysis variances to summarize major sources of variation.
• Visualization and Reporting: Produces plots and summary tables for counts distribution, mapped read density, PCA variances, hierarchical relationships, and correlation analyses.

Scientific Applications:

• Batch-effect detection and mitigation: Detects and aids mitigation of batch effects that can confound RNA-sequencing results.
• Phenotypic group consistency assessment: Evaluates consistency across phenotypic groups or experimental conditions in transcriptomic datasets.
• Support for downstream differential expression: Improves reliability of downstream analyses such as differential gene expression by flagging biased or outlier samples.

Methodology:

Computational methods explicitly include summary statistics, counts distribution and mapped read density analyses, housekeeping gene expression analysis, PCA variance analysis, unsupervised clustering including agglomerative hierarchical clustering, and correlation analyses using Pearson and Spearman coefficients for outlier and bias identification.