Page 188 - Plant Canada 2024 Proceeding

PLANT CANADA 2024

[O166] COMPARING PHENOTYPIC SELECTION WITH GENOMIC SELECTION WHEN BREEDING
FOR NEW VARIETIES OF COMMON BEAN (PHASEOLUS VULGARIS): AN EMPIRICAL STUDY.
Robert McGee , Isabella Chiaravalotti , Marysia Zaleski-Cox , Evan Wright , Karen Cichy , Diego Jarquin
1
1
3
1
2
1 1
4
D , and Valerio Hoyos-Villegas . McGill University, Department of Plant Science, Raymond Building,
21111 Lakeshore Road, Ste. Anne de Bellevue, Quebec, H9X 3V9, Canada; Michigan State University
2
(MSU), Department of Plant, Soil and Microbial Sciences, 1066 Bogue Street, East Lansing, 48824,
Michigan, USA; USDA-ARS, 1066 Bogue Street, Michigan State University (MSU), East Lansing, 48824,
3
Michigan, USA; and University of Florida (UF/IFAS), Agronomy Department, 2089 McCarty Hall B,
4
Gainesville, Florida, USA.
Correspondence to: valerio.hoyos-villegas@mcgill.ca

Since the 1900s, yield of common bean (Phaseolus vulgaris) has only increased by 1 % per year (i.e.
genetic gain), far below the ~3 % genetic gain required to feed the growing world population. To address
this challenge, new approaches are needed to rapidly develop new high yielding common bean varieties.
Breeding for yield is however challenging because it is a complex quantitative trait controlled by multiple
genes each with minor allelic effects, unlike simple traits that are controlled by small set of genes.
Genomic selection (GS) seeks to overcome this challenge by capturing the minor allelic effects of multiple
genes across the whole genome simultaneously, thereby enabling more accurate selection for
quantitative traits. Whether GS is superior to the conventional approach of selecting the highest yield
lines (phenotypic selection: PS) in common bean is unknown because no known empirical studies have
directly compared the breeding outcomes of GS and PS.

In this project, to directly compare GS and PS, two breeding pipelines were initiated in parallel from the
same pool of 38 advanced cranberry bean breeding lines. To initiate the PS pipeline, five parents were
selected solely on their phenotype (highest yielding) for 10 bi-parental crosses. The resulting F2 plants will
undergo PS in the field starting in summer 2024. In contrast, in the GS pipeline, five parents were
selected with the highest genomic estimated breeding values (GEBVs) calculated from a GS model
(rrBLUP) trained using phenotypic (seed yield) and genomic data (8,757 SNPs) from a 2021 training
population consisting of 119 advance lines (286 genotypes) from a mix of common bean market classes.
Hybridity of the resulting 212 F1 seeds from the GS pipeline was determined using newly developed
Kompetitive Allele Specific PCR (KASP) markers, revealing how the % of true hybrids differed wildly
between different parental combinations. F2 seeds from these true hybrids were advanced in the
greenhouse via single seed descent (SSD) to the F4 generation in 2023-2024. The resulting F5 seeds will
be bulked in the field in summer 2024. In summer 2026, once both pipelines, PS and GS, have reached
PYTs, genetic gain, selection accuracy, and cycle time will be directly compared to determine if GS does
or does not outperform PS in common bean. The findings from this project will help inform common bean
breeding programs of the merits and challenges of implementing GS within their existing breeding
programs.

[O167] A MULTISPECIES AMPLISEQ APPROACH TO ASSESS INTRA- AND INTER-SPECIFIC
DIVERSITY OF SPHAGNUM AND ASSIST RESTORATION EFFORTS. Mélanie Bourque , François-
1
1
Olivier Hébert and David L. Joly . Département de biologie, Université de Moncton, 18 avenue
1 1
Antonine-Maillet, Moncton, NB, Canada, E1A 3E9
Correspondence to: david.joly@umoncton.ca

Peatlands cover about 2% of the total land area of New Brunswick, and the province is the leading
producer of peat in Canada. Peatlands primarily consist of mosses from the genus Sphagnum, which
includes 160 species worldwide that are challenging to identify in the field or without substantial expertise.
After peat extraction, peatlands can be restored, and their ecosystem functions (including carbon
accumulation) can be reinstated within 10-20 years. Although Sphagnum has the ability to recover from
residual fragments and spread by spores over long distances, success of recolonization varies among
species due to hydrological changes, loss of original soil characteristics, and the unique niches each
species inhabits. Reintroduction of target Sphagnum species may be crucial to restore Sphagnum
diversity and to accelerate the re-establishment of a natural vegetation. In this study, we used a
bioinformatic approach to identify 526 regions from distinct single-copy genes exhibiting higher
interspecific diversity. We then developed an amplicon sequencing approach to quickly characterize
these regions in 700+ individuals collected from natural and restored peatlands in New Brunswick, which

187