Kastor: a reference-based comparative approach for assessment and correction of gene-fragmenting errors in long-read assemblies of small genomes

Table 1 Number and types of errors detected by Kastor

Organism	Reference set¹	Detected Errors
Organism	Reference set¹	Total	Insertions	Deletions	Substitution²
**Pseudomonas syringae508**	Set 1	2900	900	1708	900
**Pseudomonas syringae508**	Set 2	657	36	263	358
**Pseudomonas syringae508 (DFM)**³	Set 1	1674	187	559	928
**Pseudomonas syringae508 (DFM)**³	Set 2	1066	28	98	940
**Pseudomonas syringae642**	Set 1	195	71	124	-
**Citrobacter koseriMINF_9D**	Genus	951	225	209	517
**Citrobacter koseriMINF_9D**	Species	97	9	75	13
*Mycobacterium tuberculosis*	Species	18	3	3	12
*Plasmodium falciparum*W2	Species	40,681	20	40,661	-

1 – Reference data set used for Kastor correction depending on taxonomic closeness to the target assembly (i.e., reference genomes from Citrobacter species were included at the genus level, but only Citrobacter koseri genomes at the species level. See Supplementary File S4 for Reference set composition
2 – Substitutions were excluded when raw reads were unavailable
3 – basecalled using Dorado and assembled using Flye and Medaka (see Methods)

ISSN: 1471-2164