Skip to main content


Table 2 List of M. bovis BCG strains for which high per-bp coverage complete genomes are available

From: Reference genome and comparative genome analysis for the WHO reference strain for Mycobacterium bovis BCG Danish, the present tuberculosis vaccine

Strain Method type DU2 DU2 resolved Accession numbers Reference Year of publication
BCG Pasteur 1173P2 Sanger sequencing (ABI 3700) of a pUC19, two pMAQ1b, a M13 and a cloned shotgun library, earlier analysis of BAC clones had already identified the DU1 and DU2 [13] IV yes PRJEA18059, AM408590 [11] 2007
BCG Tokyo 172 SOLiD (Agencourt Bioscience Corporation) sequencing of pAGEN vector library (about 4 kb and 800 bp inserts) and a fosmid insert library (about 40 kb) I yes (2x) PRJDA31211, AP010918 [9] 2009
BCG Moreau RDJ Sanger (ABI 3730) sequencing of 2 pBluescript libraries, gap closure via PCR I no PRJEA70285, AM412059 [43] 2011
BCG Tice ATCC 35743 Illumina genome sequencing, multiplex PCR and primer walking IV no PRJNA63839, CP003494 [7] 2011
BCG Mexico 1931 Roche 454 pyrosequencing, Sanger sequencing of fos-end sequences of 250 fosmids, sequencing of three fosmids and 110 PCR end reads IV yes PRJNA45811, CP002095 [44] 2011
BCG Korea 1168P Roche 454 (GS-FLX) and Illumina (HiSeq) sequencing, gap closure via PCR and primer walking IV yes (2x) PRJNA170028, CP003900 [45] 2013
BCG Russia 368 Roche 454 (GS Junior) sequencing on a shotgun and a 3 kb paired-end library, gap closure via PCR and Sanger sequencing I yes (2x) PRJNA256163, CP009243 [46, 47] 2014
BCG 3281 (clinical isolate) Roche 454 (GS-FLX) and Illumina (Hiseq2500) sequencing, gap closure via PCR III yes PRJNA251957, CP008744 [27] 2015
BCG Russia BCG-1 Roche 454 (GS-FLX) sequencing of a shotgun library, Ion Torrent PGM sequencing of a mate-pair library I yes (2x) PRJNA306822, CP013741 [48] 2016
BCG Danish 1331 (07/270) PacBio (RSII) (long read) sequencing (235x coverage) and Illumina (MiSeq) (short read) sequencing III yes PRJNA494982, CP039850 this study 2019
  1. For each strain, we indicated the used method to create the assembled genome, the type of DU2 present in the strain, whether the DU2 was resolved in the genome assembly, the BioProject and genome assembly accession number, the reference to the study in which genome assembly method was published and the year of publication. In the ‘method’ description, we have put labor/capital-intensive aspects in bold, illustrating that our approach using solely massive parallel sequencing, is the only one that provides both high per-bp accuracy (allowing for SNP calling) and complete resolution of the assembly across large repeat regions