Related Experiment Video
Updated: Mar 9, 2026
10:00
An Ultrahigh-throughput Microfluidic Platform for Single-cell Genome Sequencing
Published on: May 23, 2018
18.4K
acdc - Automated Contamination Detection and Confidence estimation for single-cell genome data
Markus Lux1, Jan Krüger2, Christian Rinke3
1Computational Methods for the Analysis of the Diversity and Dynamics of Genomes, Bielefeld University, Universitätsstr. 25, Bielefeld, 33615, Germany. mlux@techfak.uni-bielefeld.de.
BMC Bioinformatics
|December 22, 2016
Summary
Contamination in single-cell sequencing poses a challenge, but acdc offers a novel solution. This tool uses advanced machine learning for reliable, reference-free contaminant detection and removal, improving genome assembly quality.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Single-cell sequencing is hampered by foreign DNA contamination.
- Current methods rely on reference-based detection, limiting contaminant identification.
- A reference-free approach is crucial due to fragmented genomic coverage across species.
Purpose of the Study:
- To introduce acdc, a novel tool for genomic sequence data quality control.
- To develop a reliable method for detecting both known and de novo contaminants.
- To provide a reference-free solution for contaminant identification in single-cell genomics.
Main Methods:
- Combines supervised (16S rRNA gene prediction, ultrafast alignment) and unsupervised (machine learning, dimensionality reduction, clustering) approaches.
- Employs bootstrapping for statistically sound confidence values in contaminant detection.
- Features an interactive user interface and a command-line application for workflow integration.
Main Results:
- Acdc accurately and quickly identifies contamination in diverse sequencing projects.
- The tool successfully detects both known and novel (de novo) contaminants.
- Unsupervised methods enable the identification of contaminants lacking close reference species.
Conclusions:
- Acdc reliably detects contamination in single-cell genome data.
- It complements existing tools by offering unsupervised detection of de novo contaminants.
- Acdc has the potential to significantly reduce resource expenditure in quality assurance for rapidly growing single-cell genomics data.
