Comparing SNP-based genetic linkage and physical maps in guava (Psidium guajava']java)

Genome mapping is a simplified representation of molecular markers or nucleotide sequences in chromosomes; developing accurate and dense maps is crucial for marker-assisted selection. We developed and compared genetic linkage maps obtained using JoinMap 4.0 and GACD with a physical map obtained using BLAST analysis based on Eucalyptus SNPs transferred to guava, Psidium guajava (Myrtaceae) - to serve as a reference for trait mapping in this crop. Genotyping was conducted on 112 individuals from an experimental cross between a well-known commercial cultivar and an exotic genotype (Pedro Sato x Purple guava), using the Euchip60K SNP chip, version 2.0 (72,202 SNPs); 79% of the SNPs were monomorphic. After data filtering, 1120 markers were used for map construction. The JoinMap 4.0 linkage map had 203 markers, spanning 1405.2 cM, with an average marker distance of 7.7 cM. The GACD linkage map had 186 markers and spanned 1392.7 cM, with an average marker distance of 8.8 cM. JoinMap and GACD disagreed on the estimated distances and SNP ordering. GACD showed a greater limitation than JoinMap 4.0 as it ordered markers according to their parental origin. The physical map developed using BLAST consisted of 694 hits (e-values from 8xE(-10) to 1.15xE(-26)), spanning 434.88 Mb, with an average marker interval of 0.62 Mb. Both linkage maps showed linkage groups with segments from several chromosomes compared to the physical map, indicating limitations. These results highlight the effectiveness of physical mapping through BLAST to overcome linkage mapping limitations, such as in marker grouping and ordering. The physical maps proposed here can serve as a reference for mapping and QTL estimates in guava.