Mclumi Homepage
Mclumi is a toolkit developed by adopting Markov clustering (MCL) network-based algorithms for precisely localizing unique UMIs that thus correct UMI errors. Mclumi is a Python implementation based on object-oriented programming (OOP) with interaction between modules through objects. It provides a collection of modules, including 4 main modules, dedup_basic, dedup_pos, dedup_gene, and dedup_sc_ and 1 addition module trim. Please see details here. For UMI deduplication, each main module includes 7 algorithms unique, cluster, adjacency, directional, mcl, mcl_ed, and mcl_val that take as input a bam file and output a deduplicated bam file and another 2 summary files. Every module in Mclumi can be run internally (Python inline) or externally (CLI).
__ __ ____ _ _ _ __ __ ___ _____ _ _ _ _
| \/ |/ ___| | | | | | \/ |_ _| |_ _|__ ___ | | | _(_) |_
| |\/| | | | | | | | | |\/| || | | |/ _ \ / _ \| | |/ / | __|
| | | | |___| |__| |_| | | | || | | | (_) | (_) | | <| | |_
|_| |_|\____|_____\___/|_| |_|___| |_|\___/ \___/|_|_|\_\_|\__|
Features
Clustering based on edit distance
Flexible and extendable
Dispering: increase the number of subcomponents
Shrinking: decrease the number of subcomponents
Parameterized control
More accurate for quantification esimate of molecules
Changelogs
[v0.0.4] Adding documentation to Mclumi.
[v0.0.3] A stable version of Mclumi.
[v0.0.2] Codes linked to readthedocs.
[v0.0.2] A test version of Mclumi.
Contributors
Jianfeng Sun, NDORMS, at the University of Oxford
Adam Cribbs, NDORMS, at the University of Oxford