De novo assembly of the complex genome of Nippostrongylus brasiliensis using MinION long reads

Table 5 Read mapping statistics

Read category	WGA		Unamplified
Assembly	WGA	Final	WGA	Final
Read count	1,362,775	1,362,775	2,137,475	2,137,475
Mapped	1,106,671	1,233,514	1,525,846	1,922,031
Well mapped¹	571,180	727,433	563,302	1,386,713
Uniquely mapped²	16,977	143,820	16,568	412,753
Median mapped read proportion³	91.1%	94.1%	65.6%	96.7%
Median mapped proportion excluding well-mapped reads	56.1%	65.9%	31.1%	67.2%

The number of reads that were generated from the WGA DNA and that mapped to either the assembly made from those reads (WGA) or from those generated from unamplified DNA (Final) are shown in columns 2 and 3. The number of reads that were generated from unamplified DNA (Unamplified) and that mapped to either the assembly made from the WGA reads (WGA), or from those generated from unamplified DNA (Final) are shown in columns 4 and 5. Since a similar fraction of WGA and unamplified reads do not map to the final assembly (approximately 10%), it appears that contaminating DNA was effectively removed (see ‘Methods’ for details). While the figure for the mapping of reads from the unamplified DNA to the WGA assembly is comparatively low, it is higher than the equivalent comparison of the completeness of genomes (100 × [completeness WGA/completeness final]), which is closer to 55% for Nanopolished assemblies (not shown), or 32.4% by stringent pairwise mapping. The difference is explained by redundant mapping of haplotype-specific and repeat-containing reads from the unamplified DNA
WGA whole-genome amplification
¹Reads that map at least 90% of their length to the reference assembly
²Reads that map only to the target assembly, and not the other assembly
³Median proportion of individual reads that aligned to the reference assembly

ISSN: 1741-7007