TRUmiCount

TRUmiCount corrects molecule-counting biases in next-generation sequencing by modeling PCR amplification and sequencing to provide accurate unique molecular identifier (UMI)-based molecule quantification.

Key Features:

• Correction for Amplification Bias: Distinguishes genuine UMIs from phantom UMIs and PCR chimeras to reduce overestimation caused by amplification artifacts.
• Loss Estimation: Estimates the number of molecules lost during sequencing to address underestimation from undetected molecules.
• Mechanistic Model-Based Approach: Uses a mechanistic model of PCR amplification and sequencing parameterized by PCR efficiency and sequencing depth, with parameters estimable from experimental data without calibration or spike-ins.
• Stochastic Properties Capture: Models the stochastic properties of amplification and sequencing to more accurately reflect true molecule counts.
• High Accuracy in Single-Cell RNA-Seq: Provides improved accuracy over raw UMI counts, particularly for single-cell RNA-Seq where many UMIs are typically sequenced only once.

Scientific Applications:

• Quantitative NGS experiments: Improves molecule quantification in quantitative next-generation sequencing workflows.
• RNA-Seq: Enhances accuracy of transcript abundance estimates in bulk RNA-Seq experiments.
• Single-cell transcriptomics: Refines molecule counts for single-cell RNA-Seq analyses.
• Gene expression profiling: Provides more reliable molecule counts for downstream gene expression studies and other molecular biology research.

Methodology:

Implements a mechanistic model of PCR amplification and sequencing parameterized by PCR efficiency and sequencing depth; estimates these parameters from experimental data without spike-ins or calibration, filters out phantom UMIs (PCR chimeras), and estimates molecule loss to correct raw UMI counts.