Estimating and correcting index hopping misassignments in single-cell RNA-seq data.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Index hopping is a significant source of read assignment errors in multiplexed sequencing libraries.
• This problem is exacerbated by high-capacity sequencers and single-cell RNA sequencing (scRNA-seq).

Purpose of the Study:

• To estimate the error rate of sample assignment caused by index hopping in deep, plate-based scRNA-seq.
• To develop a computational method to correct for index hopping in scRNA-seq data.

Main Methods:

• Performed deep, plate-based scRNA-seq on mouse skin cells.
• Used differential expression analysis to identify cell-type-specific signature genes.
• Quantified misassigned reads by detecting signature genes in non-native cell types.
• Developed a correction method by subtracting a fixed percentage of average expression per gene.

Main Results:

• Identified four distinct cell types from 1152 single cells.
• Estimated that 0.65% of reads per gene were misassigned across the dataset.
• The correction method increased cell-cell clustering distance and altered intermediate state assignments.

Conclusions:

• Index hopping misassignments are quantifiable and impact experimental interpretation.
• A straightforward method was developed to estimate and correct index hopping rates using distinct cell types.
• This approach is applicable to any barcoded, multiplexed scRNA-seq library for improved biological analysis.