The identification of contaminating sequences in a de novo assembly is challenging because of the absence of information on the target species. For sample types where the target organism is impossible to isolate from its matrix, such as endoparasites, endosymbionts and soil-harvested samples, contamination is unavoidable. A few post-assembly decontamination methods are currently available but are based only on alignments to databases, which can lead to poor decontamination.We present a new decontamination method based on a hierarchical clustering algorithm called MCSC. This method uses frequent patterns found in sequences to create clusters. These clusters are then linked to the target species or tagged as contaminants using classic alignment tools. The main advantage of this decontamination method is that it allows sequences to be tagged correctly even if they are unknown or misaligned to a database.Scripts and documentation about the MCSC decontamination method are available at .: data are available at Bioinformatics online.


  1. A new method for decontamination of de novo transcriptomes using a hierarchical clustering algorithm.
    Cite this
    Lafond-Lapalme J, Duceppe MO, Wang S, Moffett P, Mimee B, 2017-05-01 - Bioinformatics (Oxford, England)


  1. Joël Lafond-Lapalme

    Département de biologie, Université de Sherbrooke, Canada

  2. Marc-Olivier Duceppe

    Agriculture and Agri-Food Canada, Saint-Jean-sur-Richelieu, Canada

  3. Shengrui Wang

    Département d'informatique, Université de Sherbrooke, Canada

  4. Peter Moffett

    Département de biologie, Université de Sherbrooke, Canada

  5. Benjamin Mimee

    Agriculture and Agri-Food Canada, Saint-Jean-sur-Richelieu, Canada

Community Ratings

UsabilityEfficiencyReliabilityRated By
0 user
Sign in to rate
Tool TypeApplication
TechnologiesPerl, R
User InterfaceTerminal Command Line
Download Count0
Submitted ByBenjamin Mimee