Draft
Construction of a genetic linkage map and QTL analysis in
bambara groundnut (Vigna subterranea (L) Verdc.)
Journal: Genome
Manuscript ID gen-2015-0153.R1
Manuscript Type: Article
Date Submitted by the Author: 17-Mar-2016
Complete List of Authors: Ahmad, Nariman; Sulaimani Univerity, Department of Crop Science, Faculty of Agriculture Redjeki, Endah ; Muhammadiyah University, Faculty of Agriculture Ho, Wai ; Crops For the Future, Breeding Biotechnology and Seed systems Aliyu, Siise; The University of Nottingham, Malaysia Campus, Bioscience; Crops For the Future, BamYIELD; CSIR-Savannah Agriculture Research
Institute , Breeding Mayes, Katie; The University of Nottingham, UK campus, (c/o Sean Mayes) Bioscience Massawe, Festo; University of Nottingham Malaysia Faculty of Science, Biosciences Kilian, Andrzej ; Diversity Array Technology Pty Ltd., Director Mayes, Sean; The University of Nottingham, UK campus, Bioscience; Crops For the Future, Breeding Biotechnology and Seed systems
Keyword: Bambara groundnut, breeding, genetic mapping, QTL analysis
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Construction of a genetic linkage map and QTL analysis in bambara
groundnut (Vigna subterranea (L.) Verdc.)
Nariman Salih Ahmad1, Endah Sri Redjeki
2, Wai Kuan Ho
3,4, Siise Aliyu
3,4,5, Katie Mayes
6,
Festo Massawe3, Andrzej Kilian
7 and Sean Mayes
4,6
1. Crop Science Department, Faculty of Agricultural Sciences, Sulaimani University,
Kurdistan- Iraq.
2. Faculty of Agriculture, Muhammadiyah University, Gresik, Indonesia.
3. University of Nottingham Malaysia Campus, Jalan Broga, 43500 Semenyih, Selangor,
Malaysia.
4. Crops For the Future, Jalan Broga, 43500 Semenyih, Selangor, Malaysia.
5. CSIR-Savannah Agricultural Research Institute, Nyankpala N/R, Ghana.
6. University of Nottingham, Plant and Crop Sciences Division, Sutton Bonington Campus,
Loughborough, Leicestershire LE12 5RD, UK.
7. Diversity Array Technology Pty Ltd., Building 3, Level D, University of Canberra,
Kirinari St. Bruce, ACT2617, Australia.
Corresponding author email: [email protected]
Page 1 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
2
Abstract
Bambara groundnut (Vigna subterranea (L.) Verdc.) is an indigenous underutilised legume
which has the potential to improve food security in semi-arid Africa. So far, there are a lack of
reports of controlled breeding populations that could be used for variety development and
genetic studies. We reported here the construction of the first genetic linkage map of bambara
groundnut using a F3 population derived from a ‘narrow’ cross between two domesticated
landraces (Tiga Nicuru and DipC) with marked divergence in phenotypic traits. The map
consists of 238 DArT array and SSR based markers in 21 Linkage Groups (LGs) with a total
genetic distance of 608.3 cM. In addition, phenotypic traits were evaluated for a Quantitative
Trait Loci (QTL) analysis over two generations. A total of 36 significant QTLs were detected
for 19 traits. The phenotypic effect explained by a single QTL ranged from 11.6% to 49.9%.
Two stable QTLs were mapped for internode length and growth habit. The identified QTLs
could be significant for marker-assisted selection (MAS) in bambara groundnut breeding
programmes.
Key words: bambara groundnut; breeding; genetic mapping; QTL analysis
Page 2 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
3
Introduction
Agriculture is confronted with the twin challenges of population expansion (including a
fast pace of urbanisation) and climate change (extreme weather events, salinization,
desertification); for this reason, safeguarding food security still remains one of humanities
greatest challenges (Challinor et al. 2007; FAO 2010; Gilland, 2002). Humanity’s reliance on
essentially three major crops- rice [Oryza sativa (L.)], maize [Zea mays (L.)] and wheat
(Triticum spp.) for up to 70% of its calories is being recognised as a risky path for global food
security (Mayes et al. 2011). In addition, the gradual reduction in genetic diversity and/or
narrowing of the genetic base of these major crops during more recent breeding also represents
another dimension to the problem. For this reason, the idea that we might tap into the genetic
resources from ‘underutilised’ crops (also termed ‘neglected’ and ‘minor’) is gradually gaining
acceptance among the research community and agricultural policy think tanks (Jaenicke and
Höschle-Zeledon 2006, Mayes et al. 2011). One such important underutilised indigenous
African crop species which could make a positive contribution to global food security
(particularly in semi-arid Africa) is bambara groundnut [Vigna subterranea (L.) Verdc.].
Bambara groundnut belongs to the Leguminosae family (subfamily Papilionoideae) and has 11
pairs of chromosomes (2n=2x=22; Heller et al. 1995).
The contribution this crop could make to global food security has been previously reported
(Basu et al. 2007a; Massawe et al. 2005, 2007). The crop is reported to have drought tolerance
and the ability to adapt to marginal soils (Collinson et al. 1996; Mwale et al. 2007a, 2007b),
coupled with reasonable yield potential (BAMFOOD 2002). As a legume, it provides nitrogen
fixation for enhanced soil fertility within the agricultural system and nodulation is reported to
show good tolerance to soil nitrate (NO3-) (Dakora, 1998), with the seed providing balanced
Page 3 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
4
nutrition (Brough and Azam-Ali 1992; Mazahib et al 2013; Nti 2009). Despite these beneficial
traits, bambara groundnut has suffered some neglect within the research community. Hitherto,
there were no controlled cross breeding populations that could be used for variety development
and genetic analysis (Basu et al. 2007a). Landraces remain the main source of planting material
used by farmers (Basu et al. 2007a; Massawe et al. 2005, 2007). In an effort to develop
improved genotypes, key breeding objectives for bambara groundnut have been reported (Aliyu
et al. 2015; Massawe et al. 2005, 2007). A range of molecular marker systems have been
developed and applied to bambara groundnut landraces as a means of assessing breeding
systems, diversity and population origins (Massawe et al. 2002; Olukolu et al. 2012; Somta et al.
2011). The significance of molecular markers in speeding up breeding programmes through the
use of linkage maps and marker assisted selection (MAS) techniques is well established and
routinely practised in breeding programmes (Collard et al. 2005; Collard and Mackill 2008).
Comprehensive genetic maps have been constructed in legumes such as chickpea [Cicer
arietinum (L.)], pigeonpea (Cajanus cajan) and peanut [Arachis hypogaea (L.)] (Hong et al.
2010; Thudi et al. 2011; Saxena et al. 2012). The ability to develop genetic maps that could aid
MAS in bambara groundnut breeding programmes is a strategic objective.
We report here the construction of the first genetic linkage map of bambara groundnut
(using DArT array and SSR markers) from the progeny of two phenotypically contrasting
domesticated parental lines, alongside QTL analysis of important traits.
Materials and methods
Plant material and the development of the segregating population
Single plants of two landraces with contrasting features for growth habit (plant
morphology) and seed eye patterns, Tiga Nicuru and DipC (Fig. 1) were used as parents for
Page 4 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
5
controlled crossing in order to establish the mapping population. Based on our observation, Tiga
Nicuru from Mali has a bunchy plant morphology with an average petiole to internode ratio (P/I)
of 9 while DipC collected from Botswana has a semi-spreading morphology with an average P/I
ratio of 13. A total of 73 lines in an F2 population were obtained from this cross (DipC x Tiga
Nicuru) and advanced to F3.
Experimental set up and conditions
The F2 population was planted in the glasshouses at Sutton Bonington Campus, University
of Nottingham, UK (GPS: +52.8214, -1.2497) during the summer of 2003. Seeds were planted
directly into the soil beds at a planting distance of 25 x 25cm. Day-length was maintained at 12
hours with day and night temperatures of 28°C/23°C, respectively. The phenotypic evaluation of
the F3 population was conducted at the glasshouses from August 2011 to January 2012 as well
as at Bungah field, Gresik, Indonesia (GPS: -7.1608, -112.6471) from May to September 2010
with four replicates for each line. The plants were sown with 40cm spacing between and within
rows in the field.
Phenotypic data collection and analysis
The standard descriptors for bambara groundnut published by International Plant Genetic
Resources Institute (2000) were used as a reference for all data collection with a few
modifications. A total of 15 phenotypic traits were evaluated in the controlled environment
glasshouse for the F2 population and 29 traits for both controlled environment glasshouse and
Indonesian field for the F3 population (Table S1), in order to study the inheritance and
segregation pattern of the morphological traits.
Anderson Darling tests were used to test for normality of the distribution of the trait data
(Stephens, 1974). Data displaying non-normal distributions were transformed to try to obtain a
normal distribution by standard approaches (such as a Box-Cox transformation) and the
Page 5 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
6
Anderson Darling test repeated. Transformed-normal traits were analysed using Interval
Mapping analysis as implemented in MapQTL6.0. Non-normal traits were analysed using the
Kruskal-Wallis Ranks test with False Discovery Rates (FDR) to determine significance. Genstat
14th
Edition statistical software and MINITAB Release 16 were used to analyse trait variation
among parental lines (t-test for normal distribution trait and Mann-Whitney for non-normally
distributed traits; p<0.05), construct residual plots, linear regression analysis and detect
significant correlations (using Pearson’s coefficient correlation analysis) among the traits
(p<0.05) under different growth conditions.
PCR amplification and SSR marker analysis
Genomic DNA was extracted from young leaf using the Dellaporta protocol (1983). A
total of 33 polymorphic primer pairs (Table S2) were optimised using a three primer system as
reported by Schuelke (2000). The universal dye-labelled tag had a sequence of 5’-CAC GAC
GTT GTA AAA CGA C-3’. The sequences of the primers used are listed in Table S2.
Amplification was carried out with the following run profile: initial denaturation at 94°C for 3
min followed by 35 cycles of 94°C for 1 min, annealing step for 1 min and 72° C for 2 min with
a final extension at 72° C for 10 min.
The PCR products were checked on agarose gel before being loaded onto a CEQTM
8000
capillary sequencer (Beckman Coulter Inc., Fullerton, USA). CEQTM
8000 Fragments Analysis
Version 8 software was used to analyse the fragment sizes of the PCR products with manual
confirmation.
DArT array marker data scoring and analysis
The bambara groundnut DArT genotyping array was developed using 94 genotypes from
22 countries based on selecting a representation across available germplasms (Singrün and
Schenkel, 2003) by Diversity Array Technology Pty Ltd, Canberra, Australia
Page 6 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
7
(www.diversityarrays.com) using genomic representations derived from PstI/AluI and PstI/TaqI
digestion (Jaccoud et al. 2001; Stadler, 2009).
Genetic map construction and QTL analysis
The JoinMap4 software (van Ooijen, 2006) was used to construct the linkage map,
comprising both SSR and DArT markers. Markers were placed into linkage groups (LGs) using
default settings and a minimum LOD threshold of 3. For each LG, marker order and genetic
distance were inferred using the regression mapping algorithm. Marker classes at each locus
were summarised for all individuals into different genotypic classes for the F2 population with
the default expected ratios of 3:2:3 and 5:3 for SSR and DArT markers, respectively.
Segregation distortion was determined through a chi-square test for goodness-of-fit (p ≤0.05).
The QTL analysis was carried out by the rank sum test of Kruskal-Wallis mapping and
Interval Mapping using MapQTL 6 software (van Ooijen 2009), depending on trait or trait-
transformed distribution. The point detected with a maximum log-of-odds (LOD) score was
determined as the most likely position of the QTL on the map for Interval Mapping and the
confidence interval of a QTL location was calculated based on one-LOD support intervals. The
genome wide significance LOD threshold was empirically determined by performing a
permutation test of 10,000 iterations. For the non-normally distributed traits, a Kruskal-Wallis
test of Marker-QTL associations was implemented using False Discovery Rate (FDR)-control
(Benjamini and Hochberg 1995; van Ooijen and Maliepaard 2001). Overlapping QTLs within
the 1 LOD drop-off confidence intervals for the trait(s) across different environments and
generations were provisionally considered as the same gene effect.
Page 7 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
8
Results
Phenotypic trait analysis
The majority of the phenotypic traits analysed showed quantitative variation in the F2 and
F3 generations. Non-normal distribution was observed for days to emergence in F2 and F3
populations in both growing conditions (glasshouse and field), as did growth habit and eye
pattern around the hilum (only recorded in glasshouse) assessed in the F3 generation. The
segregation ratio of growth habit was consistent with the ratio of 3:2:3 (χ2
= 0.95 < 5.99; 2 df; p
= 0.05) for bunch, semi-bunch and spreading, which in turn suggested co-dominant inheritance.
The qualitative eye pattern around hilum trait observed was likely to be under the control of a
single dominant gene (χ2
= 0.06 < 3.84; 1 df; p = 0.05; assessed F3 generation, indicating F2
genotype). For the leaf area trait, a non-normal distribution was observed under controlled
environment conditions (p < 0.01) but not in the field (p = 0.51). Pod no./plant, plant height,
petiole length, internode length, peduncle length, leaf area, seed length, seed weight, 100-seed
weight, biomass dry weight, and shelling percentage are the traits found to be significantly
different (p < 0.05) between the parental DipC and Tiga Nicuru lines grown in glasshouse
(Table 2A).
Pearson’s correlation coefficient analysis revealed biomass dry weight to be associated
with a number of vegetative and yield related traits including leaf no./plant, plant height, petiole
length and 100-seed weight in both populations under all growing conditions (Table S3, S4 and
S5). Interestingly, the relatively strong correlation between plant spread and biomass dry weight
(r > 0.7; p < 0.001) explained as much as 55 to 65% of the trait variation in glasshouse and field
conditions. Among the evaluated traits, seed weight and pod weight recorded the highest
positive correlation with biomass dry weight (r = 0.9, p < 0.001) in F2 and F3 populations under
Page 8 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
9
the controlled environment, which also accounted for a high percentage of trait variation (r2
=
91.3% and 89.7%). However, this correlation was not significant under field conditions,
suggesting environmental influence on these traits. A strong association (r ≥ 0.9) was observed
between pod weight and seed weight traits, as might be expected, accounting for 79.6% of
variance in the field trial.
Construction of the linkage map
A total of 33 SSR markers were used to score all 73 lines of the F3 population with an
average level of residual heterozygosity of 24.9%, which was consistent with this being an F3
population. Of the 7,680 fragments immobilised on the bambara groundnut DArT array, 236
(3.1%) were identified as polymorphic markers. As a result, a total of 269 polymorphic loci
were used to construct the genetic linkage map. A slightly lower level of distorted segregation
was found in SSR markers (24.1%; 7) than in the DArT markers (32.9%; 76) and distorted
regions were distributed across a number of linkage groups. The final map consisted of 29 co-
dominant SSR markers and 209 DArT dominant markers assigned to 21 LGs, covering a total of
608.3 cM with a density of 2.6 marker per cM (although notice the clustering on LG1; Fig. 2 &
Table 1). LG10 was the longest group, consisting of 23 markers covering 76.4 cM of the map.
QTL mapping
Genome-wide significance thresholds ranged from 2.5 to 3.1 depending on the trait, trial
and generation. Among the traits being examined, 36 significant QTLs were identified by the
QTL analysis, present on a total of 8 linkage groups (Fig. 2, Table 2A and 2B). Specifically, the
following QTLs associated with important phenotypic traits related to yield potential in bambara
groundnut are worth emphasising, particularly those found to be significant different (p < 0.05)
between parental lines; pod no./plant, seed length, seed weight and 100-seed weight. (1) Seed
Page 9 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
10
weight and pod weight: the significant QTL (LG1 33.0 cM) associated with seed weight (LOD:
2.7) and pod weight (LOD: 2.6) in the F3 under glasshouse conditions was also detected as an
indicative QTL associated with seed no./plant (LOD: 2.3), explaining 17.8% and 17.0% of the
phenotypic variation in the respective significant traits. The same genomic region was also
observed to have an indicative QTL for pod no./plant (LOD: 2.3) and a significant QTL for node
no./stem trait (LOD: 3.3). (2) 100-seed weight: one significant QTL on LG7 (10.5 cM) was
found to be associated with 100-seed weight under controlled environment conditions in the F3
population.
Additionally, the following QTLs associated with phenotypic traits related to
‘domestication syndrome’ in bambara groundnut are worth highlighting. Despite both DipC and
Tiga Nicuru being domesticated lines, they differed significantly (p < 0.05) in terms of plant
height, petiole length, internode length, peduncle length, leaf area, biomass (dry weight) and
shelling percentage, with the earlier traits contributing to bunchy and semi-spreading growth
habits. (1) Plant spread: two significant QTLs adjacent to each other on LG4 were found to be
associated with this phenotypic trait in the F2 (33.5 cM; LOD 3.2) and F3 (0.0 cM; LOD 3.9)
under glasshouse conditions. (2) Growth habit: non-parametric mapping of the trait in the F3
generation in the glasshouse and the field detected a very strong association (p < 0.0005) on
LG4 0.0 cM, which was also identified as a significant QTL for plant spread (LOD: 3.9) and
double seeded pods/plant (LOD: 3.3). Another QTL on LG18 5.1 cM was detected to be linked
to the growth habit trait (p > 0.01) under field conditions, which was also an indicative QTL for
pod no./plant trait (LOD: 2.4). (3) Internode length: Data analysis for both glasshouse and the
field detected a major QTL for internode length which mapped on LG4 3.0 cM, scoring high
LOD values of 7.9 and 7.1, respectively, accounting for as much as 43.5% and 40.9% of total
Page 10 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
11
phenotypic variation. In addition, it was also identified to be strongly linked to the growth habit
trait (p < 0.0001) in the field F3 population.
Discussion
Population polymorphism, phenotypic variability and genetic mapping
In this study, polymorphic DArT Array and SSR markers were utilised to construct an
initial genetic linkage map of bambara groundnut using an F3 population from an intraspecific
cross. To the best of our knowledge, this is the first linkage map of bambara groundnut between
genotypes drawn from domesticated landraces. The two parental genotypes originating from
divergent agro-ecological backgrounds [DipC from Botswana (Southern Africa) and Tiga
Nicuru from Mali (West Africa)] have marked phenotypic differences in terms of plant
morphology and phenology (Fig 1). Furthermore, a previous diversity analysis using co-
dominant SSR and dominant DArT array markers revealed that the two parental landraces that
the specific genotypes were drawn from belong to different genetic clusters (Molosiwa et al.
2015; Stadler, 2009). A polymorphism level of 36.3% (among 124 within-species SSR markers
tested) was recorded between the parental materials, DipC and Tiga Nicuru. This is higher than
the 19% polymorphism reported by Somta et al. (2011) in their genetic diversity study of
bambara groundnut. Worthy of note is the fact that while Somta et al. (2011) used a relatively
large number of accessions (240) from diverse geographical backgrounds. Out of the 188 SSR
markers tested polymorphic in bambara groundnut accessions in their study, only 10 were
derived from within the species with the remaining from adzuki bean, cowpea and mung bean.
This could probably have accounted for the lower levels of polymorphism recorded in their
report, despite a far larger number of accessions being analysed. The level of polymorphism
observed for the DArT array markers was significantly lower (3.6%), although in line with other
Page 11 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
12
reports. For example, the polymorphism rate detected by DArT array markers ranged from
0.65% to 9.38% in six intra-specific chickpea crosses with higher rates observed (3.98% to
17.21%) in five inter-specific crosses (Thudi et al. 2011). Nevertheless, the bambara groundnut
DArT array markers had a good average polymorphic information content (PIC) value of 0.32
(Stadler 2009).
Together with the polymorphic DArT markers, the generated map consisted of 21 LGs
with a total of 608.3 cM in length. Nevertheless, the high marker-marker linkage (238 out of 269
markers in groups of at least two markers) at 89% of all markers might suggest a more
comprehensive coverage. This could be due to the parental dissimilarity which has suppressed
recombination or the developed markers clustering into particular regions of the plant genome.
In terms of the observed clustering of DArT markers such as on LG01, there could be a few
possibilities. This may indicate the presence of gene-rich regions, potentially as a reflection of
hypomethylated regions of the restriction enzyme sites, which is consistent with the
observations found in the genetic maps of other species such as chickpea and rapeseed (Raman
et al. 2013; Thudi et al. 2011). Mapping of DArT array markers to the Eucalyptus reference
genomes using the unique sequence tag of each marker has suggested that PstI-based DArT
markers are predominant at the low copy gene-rich regions (Petroli et al. 2012). Another
explanation for clustering could be the localised proliferation of a repetitive sequence on LG01
in one of the parents, but not the other (Stadler, 2009). Alternatively, the introgression of a
segment of distantly related genome into one of the parents might lead to high DArT array
marker polymorphism and clustering. In an analysis of a wheat cross between two parents
differing for the 1B-1R translocation from Rye, it was noted that very high levels of clustered
polymorphism existed in the region of the introgression (S. Mayes pers. comm, data not
published). The addition of further markers could improve the current map by reducing the
Page 12 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
13
number of linkage groups to 11 to correspond to the number of chromosomes in the bambara
groundnut genome. A larger population size is required to determine the marker order with a
greater confidence, which is one limitation of the current study. However, it is worth noting that
this is one of the first populations developed for bambara groundnut. It has been suggested that
uniformly distributed loci every 10cM over the entire genome is effective in MAS and QTL
identification (Stuber et al. 1999). Against this backdrop, the current map is suitable for MAS
and QTL analysis.
QTL for yield determinant components and ‘domestication syndrome’ in bambara
groundnut
This is also the first report of a QTL analysis for phenotypic traits in bambara groundnut. A total
of 36 significant QTLs were revealed to be associated with 19 out of 29 assessed traits. The
majority of these were located on LG1, LG4 and LG12 with the QTL detected from interval
mapping analysis explaining between 11.6 to 49.9% of the phenotypic variation of the evaluated
traits. While our data suggested a strong environmental component in most of the traits, the QTL
for internode length (3.0 cM LG4; LOD 7.9 and 7.1) and growth habit (0.0 cM LG4; p <
0.0005) were stably expressed in F3 populations evaluated in both controlled environment and
field. Previous phenotypic evaluation (analysis of variance and/or principal component analysis)
has reported a high level of variation in internode length among landrace populations (Aliyu and
Massawe 2013; Molosiwa et al. 2002). Consistent with this, the stable QTL detected in this
study explained a high proportion of phenotypic variation (more than 40%) from the F3
generation grown under controlled environment glasshouse and field conditions. This
observation of relatively strong genetic effects in the current study could be further supporting
evidence for the previous report that internode length is under the control of a single dominant
Page 13 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
14
gene for domestication and that this locus still retains variation in domesticated germplasms
(Basu et al. 2007c). It is worth noting that, although relatively weak, significant associations of
this trait with seed weight and pod weight (r = 0.3 – 0.4, p < 0.05; Table S4 and S5) were
observed in the current study under both glasshouse and field conditions which could be useful
in MAS. The regression analysis (Fig. 3) has suggested that seed weight and pod weight
accounted for 11.1% and 7.6%, respectively of the variation in internode length of F3 population
grown in glasshouse. Furthermore, this same QTL was identified in the F3 field trial to be
significantly associated with growth habit. As expected, a strong negative correlation was
observed between internode length and growth habit (r = -0.7; p < 0.001). Therefore the
progenies which inherit this QTL allele are likely to exhibit a transition from bunch to semi-
bunch growth habit. This is a favourable agronomic trait particularly in small scale and/or
subsistence mixed farming systems typical of agro-ecologies where bambara groundnut is
cultivated. Generally, the transformation from spreading type to semi-bunch/bunch type of
growth habit in bambara groundnut has been reported as one of the key ‘domestication
syndrome’ changes of the crop (Aliyu and Massawe 2013; Basu et al. 2007b). We postulate that
QTL identified in this study to be associated with internode length could be one of the key
genetic loci selected for by farmers (shorter internode length) during the course of domestication
of this crop, which has progressively resulted in a higher P/I ratio (7-9 for semi-bunch, and > 9
for complete bunchy) of landraces. This could possibly explain the observation that 47% and 8%
of the landraces in IITA accessions were semi-bunch and spreading types respectively (Goli et
al. 1995). Overall, it can be observed that a number of genes controlling vegetative growth are
co-localised from 0.0 to 33.5 cM on LG4.
It is possible that certain genetic loci exhibit pleiotropic effects. Specifically, one genetic
locus on 33.0 cM of LG1 is worth noting for a potential pleiotropic effect. This is a major QTL
Page 14 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
15
for biomass dry weight, node no./plant, pod weight and seed weight and an indicative QTL for
pod no. and seed no./plant. More than one QTL have been identified for different traits mapped
to/near the equivalent position in soybean (Wenxin et al. 2008; Zhang et al. 2004). However, we
do not rule out the possibility of having multiple tightly linked genes clustered together at this
locus in the current analysis. Further investigation using larger populations would lead to a
better understanding of the nature of this QTL and whether it is really a number of linked QTL.
While this locus could be significant for MAS related to yield potential, its environmental
dependence (detected only under glasshouse conditions) is a potentially limiting factor. For
MAS related to yield potential, we would recommend using it in conjunction with other QTLs.
Conclusion
A total of 33 polymorphic species-specific SSR markers and 236 DArT array based
markers were used to construct an initial linkage map allowing QTL analysis of important
phenotypic/agronomic traits. To the best of our knowledge, this is the first report in the literature
of genetic mapping between domesticated landraces and QTL analysis in this underutilised
species. The map comprises of 29 SSR and 209 DArT array markers grouped into 21 linkage
groups. In total, 36 significant QTL were detected using interval mapping and non-parametric
mapping. Most of the QTL detected were clustered on LG1, LG4 and LG12. Specifically, QTL
linked to important traits related to ‘domestication syndrome’ and yield potential in bambara
groundnut have been identified. The current map could be useful in a breeding improvement
programme of this underutilised crop and could allow MAS strategies to be deployed.
Acknowledgements: The authors gratefully acknowledge the EU BAMLINK project for
providing the funding for this project and MOHESR-IRAQ for awarding a scholarship to the
first author.
Page 15 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
16
References
Aremu, M.O., Olaofe O., Akintayo E.T. .2006. Chemical Composition and Physicochemical
Characteristics of Two Varieties of Bambara Groundnut (Vigna subterrenea) J. Appl.
Sci. 6:1900-1903.
Aliyu, S., Massawe, F., and Mayes, S. 2015. Beyond landraces: developing improved
germplasm resources for underutilized species- a case for Bambara groundnut.
Biotechnol. Genet. Eng. Rev (1-2) 37–41. doi: 10.1080/02648725.2014.992625
Aliyu, S., and Massawe, F. J. 2013. Microsatellites based marker molecular analysis of
Ghanaian Bambara groundnut (Vigna subterranea (L.) Verdc.) landraces alongside
morphological characterization. Genet Resour Crop Evol 60:777–787. doi:
10.1007/s10722-012-9874-y
Aykroyd, W. R., and Doughty, J. 1982. Legumes in Human Nutrition. FAO Food and Nutrition
Paper 20. FAO, Rome, Italy.
Azam-Ali S.N., Sesay A., Karikari S.K., Massawe F.J., Aguilar-Manjarrez J., Bannayan M., and
Hampson K.J. 2001. Assessing the potential of an underutilized crop: A case study using
Bambara groundnut. Exp. Agric. 37:433-472.
BAMFOOD, 2002. Increasing the Productivity of Bambara Groundnut (Vigna subterranea (L.)
Verdc.) for Sustainable Food Production in Semi-Arid Africa. Project report, INCO-DC-
International Cooperation with Developing Countries: The University of Nottingham,
Sutton Bunington, UK.
Basu S., Roberts, J.A., Azam-Ali, S.N., and Mayes, S. 2007a. Bambara groundnut. In: Genomic
mapping and molecular breeding in plants, volume 3, Pulses, sugar and tuber crops.
Edited by Kole C., Springer, New York pp 159-173
Page 16 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
17
Basu, S., Roberts J.A., Azam-Ali S.N. and Mayes, S. 2007b. Development of microsatellite
markers for Bambara groundnut (Vigna subterranea L. Verdc.) - an underutilized African
legume crop species. Mol. Ecol. Notes. 7:1326-1328.
Basu, S., Mayes, S., Davey, M., Roberts, J.A., Azam-Ali, S.N., Mithen, R. and Pasquet, R.S.
2007b. Inheritance of 'domestication' traits in Bambara groundnut (Vigna subterranea
(L.) Verdc.). Euphytica. 157: 59-68.
Benjamini, Y. and Hochberg, Y. 1995. Controlling the fase dicovery rate: a practical abd
pwoerful appraoch to multiple testing. J. R. Stat. Soc. 57: 289-300.
Brough, S.H. and Azam-Ali, S. N. 1992. The effect of soil moisture on the proximate
composition of Bambara groundnut (Vigna subterranea (L.) Verdc). J. Sci. Food. Agric.
60: 197–203.
Challinor, A., Wheeler, T., Garforth, C., Craufurd, P. and Kassam, A. 2007. Assessing the
vulnerability of food crop systems in Africa to climate change. Clim. Change. 83:381–
399.
Collard, B.C.Y., Jahufer, M.Z.Z., Brouwer, J.B., and Pang, E.C.K. 2005. An introduction to
markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop
improvement: The basic concepts. Euphytica. 142: 169-196.
Collard, B.C.Y., and Mackill, D. J. 2008. Marker-assisted selection: an approach for precision
plant breeding in the twenty-first century. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 363
(1491), 557–72. doi:10.1098/rstb.2007.2170
Collinson, S.T., Azam-Ali, S.N., Chavula, K.M. and Hodson, D.A. 1996. Growth, development
and yield of Bambara groundnut (Vigna subtarranea) in response to soil moisture. J.
Agric. Sci. Cambridge. 126: 307-318.
Page 17 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
18
Dakora, F.D. 1998. Nodule function in symbiotic Bambara groundnut (Vigna subterranea L.)
and Kersting's bean (Macrotyloma geocarpum L.) is tolerant of nitrate in the root
medium. Ann. Bot. 82:687-690.
Dellaporta, S.L., Wood J., and Hicks, B.J. 1983. A plant DNA minipreparation version II. Plant
Mol. Biol. Report. 1:19-21.
Food and Agricultural organization (FAO) (2010). The State of Food Insecurity in the World:
Addressing food insecurity in protracted crises. Rome, Italy
Gilland., B. 2007. World population and food supply, Can food production keep pace with
population growth in the next half-century? Food. Policy. 27:47–63.
Goli, A. E., Begemann, F., and Ng, N.Q. 1995. Characterization and evaluation of IITA’s
Bambara groundnut collection. In Proceedings of the Workshop on Conservation and
Improvement of Bambara Groundnut (Vigna subtarranea (L.) Verdc.). Harare,
Zimbabwe, 14-16th November. Edited by J.H., Begemann, and J. Mushonga.
Heller J., Begemann F., and Mushonga J. (Editors). 1995. Bambara groundnut. Vigna
subterranea (L.) Verdc. Promoting the conservation and use of underutilised and
neglected crops. Proceedings of the Workshop on Conservation and Improvement of
Bambara Groundnut (Vigna subterranea (L.) Verdc.), 14-16 November 1995, Harare,
Zimbabwe. Institute of Plant Genetics and Crop Plant Research, Gatersleben/Department
of Research & Specialist Services, Harare/International Plant Genetic Resources
Institute, Rome, Italy, pp. 166.
Hong, Y., Chen X., Liang X., Liu H., Zhou G., Li S., Wen S., Holbrook C.C., and Guo, B. 2010.
A SSR-based composite genetic linkage map for the cultivated peanut (Arachis hypogaea
L.) genome. BMC. Plant. Biol. 10:Article No. 17.
Page 18 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
19
Howell, J.A., Eshbaugh, W.H., Guttman, S., and Rabakonandrianina, E. 1994. Common names
given to Bambara groundnut (Vigna subterranea: Fabacea) in central madagasqar. Econ.
Bot. 48:217-221.
IPGRI/IITA/BAMNET. 2000. Descriptors for Bambara groundnut (Vigna subterranea).
International Plant Genetic Resources Institute, Rome, Italy; International Institute of
Tropical Agriculture, Ibadan, Nigeria; The International Bambara groundnut Network,
Germany. ISBN 9290434619
Jaccoud, D., Peng K, Feinstein, D., Kilian, A. 2001. Diversity Arrays: a solid state technology
for sequence information independent genotyping. Nucleic. Acid. Res. 29(4): e25.
Jaenicke, H., & Höschle-Zeledon, I. (Editors.). 2006. Strategic Framework for Underutilized
Plant Species Research and Development, with Special Reference to Asia and the
Pacific, and to Sub-Saharan Africa. Rome, Italy: International Centre for Underutilsed
Crops, Colombo, Sri Laka and Global Facilitation Unit for Underutilized Species, Rome,
Italy.
Kalia, R.K., Rai, M.K., Kalia S., Singh, R., and Dhawan, A.K. 2011. Microsatellite markers: an
overview of the recent progress in plants. Euphytica. 177:309-334.
Massawe, F.J., Dickinson, M., Roberts, J.A., and Azam-Ali, S.N. 2002. Genetic diversity in
Bambara groundnut (Vigna subterranea (L.) Verdc) landraces revealed by AFLP
markers. Genome. 45:1175-80.
Massawe, F.J., Mwale, S.S., Azam-Ali, S.N., and Roberts, J.A. 2005. Breeding in Bambara
groundnut (Vigna subterranea (L.) Verdc.): strategic considerations. African. J.
Biotechnol 4:463-471.
Massawe, F. J., Mwale, S. S., Azam-Ali, S. N., and Roberts, J. A. 2007. Towards Genetic
Improvement of Bambara Groundnut [Vigna subterranea ( L .) Verdc.]. In Breeding of
Page 19 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
20
Neglected and Under-Utilized crops: Spices and herbs (p. 468). Edited by Ochatt, S. and
Jain, S.M. Science Publishers.
Mayes, S., Massawe, F. J., Alderson, P. G., Roberts, J. A., Azam-Ali, S. N., and Hermann, M.
2011. The potential for underutilized crops to improve security of food production. J.
Exp. Bot. 63(3): 1075–9. doi:10.1093/jxb/err396
Mayes S., Basu S., Murchie E., Roberts J.A., Azam-Ali S.N., Stadler F., Mohler V., Wenzel G.,
Massawe F., Kilian A., Bonin A., Beena A., Sheshshayee M.S. 2009. BAMLINK - a Cross
Disciplinary Programme to Enhance the Role of Bambara Groundnut (Vigna subterranea L.
Verdc.) for Food Security in Africa and India. In International Symposium on
Underutilized Plants for Food Security, Nutrition, Income and Sustainable Development
806:137-149.
Mazahib, A.M., Nuha, M. O., Salawa, I. S., and Babiker, E. E. 2013. Some nutritional attributes
of Bambara groundnut as influenced by domestic processing. Int. Food. Res. J. 20:1165–
1171.
Molosiwa, O. O., Aliyu, S., Stadler, F., Mayes, K., Kilian, A., and Mayes, S. 2015. SSR marker
development, genetic diversity and population structure analysis of Bambara groundnut
[Vigna subterranea (L.) Verdc.] landraces. Genet. Resour. Crop Evol. 62(8): 1225–1243.
doi:10.1007/s10722-015-0226-6
Molosiwa, O.O. 2012. Genetic diversity and population structure analysis of Bambara
groundnut [Vigna subterranea (L.) Verdc.] landraces using morpho-agronomic characters
and SSR Markers. Ph.D. Thesis. University of Nottingham, UK.
Mwale, S.S., Azam-Ali, S.N. and Massawe, F.J. 2007a. Growth and development of Bambara
groundnut (Vigna subterranea) in response to soil moisture 1. Dry matter and yield. Eur. J.
Agron. 26: 345–353.
Page 20 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
21
Mwale, S.S., Azam-Ali, S.N. and Massawe, F.J. 2007b. Growth and development of Bambara
groundnut (Vigna subterranea) in response to soil moisture 2. Resource capture and
conversion. Eur. J. Agron. 26: 354–362.
Nti, C. A. 2009. Effects of Bambara groundnut (Vigna subterranea) variety and processing on
the quality and consumer appeal for its products. Int. J. Food. Sci. Technol. 44: 2234–2242.
Olukolu, B.A., Mayes, S., Stadler, F., Ng, N.Q., Fawole, I., Dominique, D., Azam-Ali, S.N.,
Abbott, A.G., and Kole, C. 2012. Genetic diversity in Bambara groundnut (Vigna
subterranea (L.) Verdc.) as revealed by phenotypic descriptors and DArT marker analysis.
Genet. Resour. Crop Evol. 59:347-358. doi: 10.1007/s10722-011-9686-5.
Petroli C.D., Sansaloni C.P., Carling J., Steane D.A., Vaillancourt R.E., Myburg A.A., et al.
(2012) Genomic characterization of DArT markers based on high-density linkage analysis
and physical mapping to the Eucalyptus genome. PLoS ONE 7(9): e44684.
doi:10.1371/journal.pone.0044684
Raman H., Raman R., Kilian A., Detering F., Long Y., Edwards D. et al. (2013) A consensus
map of rapeseed (Brassica napus L.) based on diversity array technology markers:
applications in genetic dissection of qualitative and quantitative traits. BMC. Genomics. 14:
277.
Saxena, R.K., Penmetsa, R. V., Upadhyaya, H.D., Kumar, A., Carrasquilla-Gracia, N.,
Schlueter, J.A., Farmer, A., Whaley, A. M., Sarma, B. K., May, G.D., Cook, D. R.,
Varshney, R. K. 2012. Large-scale development of cost-effective single-nucleotide
polymorphism marker assays for genetic mapping in pigeonpea and comparative mapping
in legumes. DNA. Reseacrh. 19: 449-461.
Page 21 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
22
Schuelke, M. 2000. An economic method for the fluorescent labelling of PCR fragments. A poor
man’s approach to genotyping for research and high-throughput diagnostics. Nat.
Biotechnol. 18: 233-234.
Singrün, C. and Schenkel, W. 2003. Fingerprinting of bambara groundnut germplasm with
molecular markers. In: Proceedings of the International Bambara Groundnut
Symposium, Botswana College of Agriculture, Botswana, 8-12 August 2003, pp 161-
173.
Somta, P., Chankaew, S., Rungnoi, O., Srinives, P. 2011. Genetic diversity of the Bambara
groundnut (Vigna subterranea (L.) Verdc.) as assessed by SSR markers. Genome. 54:
898-910. doi: 10.1139/g11-056.
Stadler, F. 2009. Analysis of differential gene expression under water-deficit stress and genetic
diversity in Bambara groundnut [Vigna subterranea (L.) Verdc.] using novel high-
throughput technologies. Ph.D Thesis, Technischen Universität München, Germany.
Stephens, M.A. 1974. EDF Statistics for Goodness of Fit and Some Comparisons. J. Am. Stat
.Assoc. 69:730-737.
Stuber, C.W., Polacco, M., and Lynn, M. 1999. Synergy of empirical breeding, marker-assisted
selection, and genomics to increase crop yield potential. Crop. Sci. 39:1571-1583.
Thudi, M., Bohra, A., Nayak S.N., Varghese, N., Shah, T.M., Penmetsa, R.V.,
Thirunavukkarasu, N., Gudipati, S., Gaur, P.M., Kulwal, P.L., Upadhyaya, H.D.,
Kavikishor, P.B., Winter, P., Kahl, G., Town, C.D., Kilian, A., Cook, D.R., and
Varshney, R.K. 2011. Novel SSR markers from BAC-end sequences, DArT arrays and a
comprehensive genetic map with 1,291 marker loci for chickpea (Cicer arietinum L.).
PLoS One 6:e27275. doi: 10.1371/journal.pone.0027275. PONE-D-11-10266 [pii].
Page 22 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
23
Van Ooijen, J.W. 2006. JoinMap 4. Software for the calculation of genetic linkage maps in
experimental populations. Kyazma BV: Wageningen, Netherlands.
Van Ooijen, J.W. 2009. MapQTL 6. Software for the mapping of quantitative trait loci in
experimental populations of diploid species. Kyazma BV: Wageningen, Netherlands.
Van Ooijen, J.W., and Maliepaard, C. 2001. MapQTL version 4.0: Software for the calculation
of QTL positions on genetic maps. CPRO-DELO, Wageningen. The Netherlands.
Vietmeyer D.N. (1979) Tropical legumes: Research for the future. National Accademic press,
Washington.
Wenxin, L., Da-Hao, Z., Kyujung, V., and Suk-Ha, L. 2008. QTL Mapping for Major
Agronomic Traits across Two Years in Soybean (Glycine max L. Merr.). Crop. Sci.
Biotechnol. 11:171-176.
Zhang, W.K., Wang, Y.J., Luo, G.Z., Zhang, J.S., He, C.Y., Wu, X.L., Gai, J.Y., and Chen, S.Y.
2004. QTL mapping of ten agronomic traits on the soybean (Glycine max L. Merr.)
genetic map and their association with EST markers. Theor. Appl. Genet. 108:1131-
1139.
Page 23 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
24
Table 1. Distribution of SSR and DArT markers among 21 LGS in the map constructed from a
F3 population of DipC x Tiga Nicuru cross.
LG Length (cM) No. of
markers
Type of marker
SSR DArT
1 72.6 46 3 43
2 36.6 11 0 11
3 38.1 15 0 15
4 33.5 13 3 10
5 74.2 28 5 23
6 20.9 13 0 13
7 13.3 12 2 10
8 16.8 5 2 3
9 1.6 3 0 3
10 76.4 23 4 19
11 15.3 14 1 13
12 47.5 10 1 9
13 18.0 9 1 8
14 3.1 9 1 8
15 42.9 7 1 6
16 8.0 5 1 4
17 23.5 4 2 2
18 34.4 4 1 3
19 30.2 3 1 2
20 0.0 2 0 2
21 1.4 2 0 2
Total 608.3 238 29 209
Page 24 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
25
Table 2A. QTL analysis from interval mapping
Traits Population/condition LG Position (cM) Nearest marker Interval mapping
LOD PTa PVE
b Additive
Terminal leaflet length (TLL) F2 CE 8 0.0 Bam2coL63 2.3
2.6 11.6 -0.6
F3 CE 8 0.0 Bam2coL63 3.1 19.7 -0.4
Terminal leaflet width (TLW) F2 CE 5 74.2 bgPt-595387 2.6 2.5 15.3 0.3
F3 CE 3 19.7 bgPt-600935 3.2 2.6 20.4 0.2
Plant spread (PS) F2 CE 4 33.5 bgPabg-597624 3.2
2.7 18.0 5.5
F3 CE 4 0.0 BN6b 3.9 24.6 3.7
Node no./stem (NN) F3 CE
1 33.0 bgPabg-596774 3.3 2.6
21.1 -1.1
4 11.2 bgPt-600898 2.7 17.9 1.1
F3 field 3 30.2 bgPabg-595707 2.8 2.7 18.4 1.0
Internode length (IL)* F3 CE 4 3.0 bgPabg-596988 7.9 2.6 43.5 0.7
F3 field 4 3.0 bgPabg-596988 7.1 2.7 40.9 0.3
Double seeded pods/plant (DPN) F2 CE 4 33.5 bgPabg-597624 3.3 2.8 19.2 0.7
F3 CE 4 0.0 BN6b 3.3 2.9 21.7 0.5
Peduncle length (PEL)* F3 CE 4 1.0 bgPt-423527 9.7 2.7 49.9 0.9
Pod weight (PWE) F3 CE 1 33.0 bgPabg-596774 2.6 2.5 17.0 -6.8
Pod length (PLE) F3 CE 12 15.0 - 4.6 2.7 28.0 0.8
F3 field 11 0.0 bgPabg-595822 3.0 2.5 19.9 0.1
Pod width (PWD) F3 CE 12 20.0 - 5.7 2.4 32.7 0.5
Pod length of double seeded (DPL)* F3 CE 1 0.0 bgPabg-597086 3.8
2.7 24.5 -1.5
12 10.5 - 3.3 21.7 1.6
Pod width of double seeded (DPW) F3 CE 12 17.0 - 4.0 2.8 24.0 0.5
Seed weight (SWT)* F3 CE 1 33.0 bgPabg-596774 2.6 2.6 17.8 -0.5
Biomass dry weight (BDW)* F3 CE 1 33.0 bgPabg-596774 3.5 3.0 22.4 -11.6
F3 field 1 28.9 bgPt-602039 2.9 2.9 17.6 -1.8
Shelling % (SH)* F2 CE 12 47.5 bgPt-595486 4.8 2.6 26.3 -4.0
F3 CE 7 13.3 bgPabg-594335 3.0 2.9 19.4 3.4
100-seed weight (HSW)* F3 CE 7 10.5 bgPt-601852 2.6 2.5 17.3 4.3
*: significantly different between parental lines; CE: controlled environment; a: permutation 10,000 times test;
b: percentage of total
phenotypic variation explained by the QTL
Page 25 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
26
Table 2B: QTL analysis from Kruskal Wallis analysis (FDR-corrected, Q < 0.05)
Traits Population/condition LG Position (cM) Locus Kruskal-Wallis analysis
K Siga
Days to emergence (DE)
F2 CE 5 74.2 bgPt-595387 7.8 ***
17 23.5 PRIMER16 12.3 ****
F3 field 1 9.3 bgPabg-423556 7.3 ***
13 9.0 bgPt-598091 6.9 ***
Growth habit (GH)
F3 CE 4 0.0 BN6b 23.4 *******
10 70.1 bgPabg-596205 8.3 ****
F3 field
4 0.0 BN6b 17.6 ******
4 3.0 bgPabg-596988 18.1 *******
14 0.0 bgPt-597832 7.4 ***
18 5.1 PRIMER10 9.7 ***
Eye pattern around hilum (EP) F3 CE 12 22.5 bgPabg-594999 29.7 *******
18 0.0 bgPabg-594261 9.3 ****
CE: controlled environment; a: Significance: ***: 0.01, ****: 0.005, *****: 0.001, ******: 0.0005, *******: 0.0001
Page 26 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
27
List of figures
Figure 1. DipC and Tiga Nicuru with contrasting growth habit and seed eye pattern were used
as parental lines to develop this segregating bi-parental population. (A) DipC plant and (B) seed;
(C) Tiga Nicuru plant and (D) seed.
Figure 2. The genetic linkage map of bambara groundnut consisting of 21 linkage groups from
the cross of DipC x Tiga Nicuru. The position of 29 SSR and 209 DArT markers are given in
centimorgan to the left of the linkage groups and the name of markers to the right with asterisk
(*) indicating markers with segregating distortion. The position of the maximum LOD value of a
particular QTL is written at the top of QTL pointer with the growth condition of the derived
QTL was indicated in black four pointed star for F2 progeny, by black rectangle for glasshouse
experiment and white rectangle for field F3 data. QTL confidence intervals (1 LOD drop-off) are
represented by plain lines (LOD score ≥ GW threshold). QTL detected with Kruskal-Wallis
analysis (KW) are discontinuous with no confidence interval. The LOD plots of internode length
(IL; LG4 F3 population grown at field) and pod width (PWD; LG12 F3 population grown in
glasshouse) are given as examples.
Figure 3. Regression plots of internode length vs (A) seed weight and (B) pod weight observed
from F3 segregating population grown under controlled environments.
Page 27 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
A C
B D
Page 28 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Page 29 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Page 30 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Page 31 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Page 32 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Page 33 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
DraftA.
B.
Page 34 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Table S1: Phenotypic traits evaluated during the field and controlled environment.
No Trait Unit Description
1 Days to emergence d Number of days from sowing to the
appearance of first true leaf on the soil
surface.
2 Days to flowering d Recorded from seedling emergence to the
appearance of the first flower(s)
3 Plant height cm Measured from the ground level to the tip
of the highest point recorded at 10 weeks
after planting
4 Petiole length mm Measured from the stem node to the
junction of the three leaflets at the longest
stem at the fourth node at 10 weeks after
planting
5 Flower no./plant - Counted each 2-3 days from the first day of
flowering for the duration of study.
6 Terminal leaflet length mm Length of median leaflet at the fourth node
recorded at 10 weeks after planting
7 Terminal leaflet width mm Width of median leaflet at the fourth node
recorded at 10 weeks after planting
8 Leaf area m2 Estimated based on the central leaflet
length and width using the method of
Cornelissen et al. (2002) in the following
equation:
Aplant = 0.86 * Leaf number [0.91 * 3
(0.95 * Length * Width * π /4)]
Where leaf number = leaf number/plant;
length and width being mean length and
width of the terminal leaflet of five
leaves/plant, and π = 3.1416.
9 Plant spread cm Widest point between two opposite points
of the plant canopy recorded at 10 weeks
after planting
10 Stem no./plant - Recorded at harvest
11 Node no./stem - Average node number on three stems/plant,
recorded at harvest.
Page 35 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
12 Internode length mm Length of fourth internode of the longest
stem, recorded at 10 weeks after planting
13 Growth habit (ratio of
petiole/internode)
- Bunch: P/I > 9; Semi-bunch: P/I =7-9;
spreading: P/I < 7)
14 Pod no./plant - Counted at harvest. Number of pods with
more than one seed was also determined.
15 Double seeded pods/plant - Total number of double seeded pods per
plant counted after harvesting
16 Peduncle length mm
17 Pod weight/plant g Weight of dried pods (at 12% moisture
content) was recorded after maintaining the
harvest pods for three weeks at 37oC
18 Pod length mm Digital Vernier Caliper (model no. OD-
15GP, serial no. 211810, Mitutoyo UK
Ltd.) was used to measure the greatest
length and width of five dried pods
containing one seed.
19 Pod width mm Digital Vernier Caliper was used to
measure the greatest length and width of
four dried seeds (at 12% moisture content).
20 Double seeded pod length mm Digital Vernier Caliper (model no. OD-
15GP, serial no. 211810, Mitutoyo UK
Ltd.) was used to measure the greatest
length of five dried pods containing two
seed.
21 Double seeded pod width mm Digital Vernier Caliper (model no. OD-
15GP, serial no. 211810, Mitutoyo UK
Ltd.) was used to measure the greatest
width of five dried pods containing two
seed.
22 Seed length mm Digital Vernier Caliper was used to
measure the greatest length of four dried
seeds (at 12% moisture content).
23 Seed width mm Digital Vernier Caliper was used to
measure the greatest width of four dried
seeds (at 12% moisture content).
Page 36 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
24 Seed no./plant - Total number of seeds per plant counted
after harvesting
25 Seed weight/plant g Average width of 5 seeds from plant
26 Biomass dry weight g Weight of above ground biomass of
harvested plants
27 Shelling % Measured as an average of all pods/plant,
based on the weight of matured dried seeds
compared to the weight of dried pods.
28 100-seed weight g Recorded after harvest at 12% moisture
content
29 Eye pattern around hilum - 0 = No eye pattern; 1 = Butterfly; 2 =
Triangular; 3 = Mottled; 4 = Thick dotted
lines; 5 = Circular; 6 = Thin lines
Page 37 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Table S2: SSR primer sequences and parental allelic scoring
Name Forward sequence (5ʹ-3ʹ) Reverse sequence (3ʹ-5ʹ) Optimal TA (oC)
Fragment length (bp)
Parent 1 Parent 2
PRIMER2 CGTGGATACCCATACCGTCT TAAGTCCATTTTGTCCGATTGA 51 171 173
PRIMER7 GTAGGCCCAACACCACAGTT GGAGGTTGATCGATGGAAAA 54 210 212
PRIMER10 TCAGTGCTTCAACCATCAGC GACCAAACCATTGCCAAACT 54 260 234
PRIMER15 AGGAGCAGAAGCTGAAGCAG CCAATGCTTTTGAACCAACA 58 238 212
PRIMER16 CCGGAACAGAAAACAACAAC CGTCGATGACAAAGAGCTTG 55 189 187
PRIMER19 AGGCAAAAACGTTTCAGTTC TTCATGAAGGTTGAGTTTGTCA 57 273 235
PRIMER26 CGCTCATTTTAACCAGACCTC CAAACAAACCAACGGAATGA 55 183 185
PRIMER32 TTCACCTGAACCCCTTAACC AGGCTTCACTCACGGGTATG 55 247 251
PRIMER37 CCGATGGACGGGTAGATATG GCAACCCTCTTTTTCTGCAC 60 258 260
PRIMER38 TCACACTTGCAATGGTGCTT TCGTTGTTTCTCTTTTCATTGC 57 194 191
PRIMER43 CTTGATGCTACCGAGAGAGAG AGGCTCCAACAATGCGATAG 55 199 205
PRIMER45 CGTGGATACCCATACCGTCT AAGTCCATTTTGTCCGATTGA 52 171 173
PRIMER48 TACCTGCATTCGGGACAGTT TTCACTCTTTCTTGATCACATGC 60 238 230
PRIMER65 GGACGTGAATCGATGGAGAT TCCTTCCCCCTTCTCTGATT 55 172 176
PRIMER66 CGTTAGATCTGAGACGCCATT CATCCATCACCTGTCACCAG 60 225 213
PRIMER85 TTTCCAGATTGGATCGTTGA TGTCTTCACACCGGAATTTG 58 248 252
PRIMER88 TGTGGTTGTGCTCCTTCTCA GGGAAGAAGAGTGAAGTTGGAA 62 233 239
PRIMER95 AAGTCCATTTTGTCCGATTGA CGTGGATACCCATACCGTCT 58 168 170
PRIMER98 TTTTGTCACTGTTTGCCACAA AGATTTATATCTGGATGAGAGAGAGAG 57 264 294
PRIMER103 AAATTCAAAGGCCTGGAAAAA TTTTTGAGTTCTGCGAGCAA 57 210 220
GH-19-B2-D9 ATCAAAATCAAGCAAATGAGA ACCTTTTACGCTCATTTTAACCAG 50 236 238
BamcoL17 AACCTGAGAGAAGCGCGTAGAGAA GGCTCCCTTCTAAGCAGCAGAACT 58 162 166
Bam2coL33 ATGTTCCTTCGTCCTTTTCTCAGC AAAACAATCTCTGCCCCAAAAAGA 54 253 255
Bam2coL63 AAAATCTCACTCGGATGGCATGTG TGGAATCACCTGATAGTAGTGTATTGG 55 293 295
Bam2coL80 GAGTCCAATAACTGCTCCCGTTTG ACGGCAAGCCCTAACTCTTCATTT 58 220 224
mBam3co7 GGGTTAGTGATAATAAATGGGTGTG GTCATAGGAAAGGACCAGTTTCTC 59 267 275
mBam3co33 TGTGTCTGTTTGTGGGGATATGTA TTATCCCGGTCCTAATTCATCTTA 58 295 319
AG81 ATTTTCCAACTCGAATTGACC TCATCAATCTCGACAAAGAATG 52 202 190
BN 6b CACTACCCTGTTCTTCATCCGT CATTGCACGTCATAGAATTTGG 53 146 150
BN 145 GGCACTGGTAGCAACGAAA CGTGGACGTAACAACACAACAC 50 150 154
BN 259 CGATTGCACGTCATAGAATTTG GTTCCAGACACTACCCTCGTTC 50 159 163
D.24269 AGGTTCATGATCGTAGATGTGGAT ACGATATCATACTGACATGTTTCATAC 60 246 238
D.35497 ACTTTTAGCTCTTGTCAGGAAACG TCTTTCTACTTTTCTCTGGCTGGT 55 168 202
Page 38 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Table S3: Correlation analysis of recorded traits in the F2 population grown in glasshouses with p-values. Traits 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Days to emergence 1
Leaf no./plant 2 -0.107
0.366
Days to flowering 3 0.609 -0.199
0.000 0.091
Plant height 4 0.275 0.575 0.217
0.018 0.000 0.065
Petiole length 5 0.125 0.673 0.200 0.749
0.301 0.000 .097 0.000
Terminal leaflet length 6 -0.029 0.183 -0.192 0.213 0.312
0.810 0.129 0.104 0.071 0.009
Terminal leaflet width 7 -0.064 0.495 -0.219 0.350 0.447 0.752
0.593 0.000 0.063 0.002 0.000 0.000
Plant spread 8 0.055 0.635 0.049 0.615 0.832 0.547 0.590
0.641 0.000 0.681 0.000 0.000 0.000 0.000
Pod no./plant 9 0.079 0.530 0.133 0.545 0.768 0.388 0.456 0.728
0.507 0.000 0.263 0.000 0.000 0.001 0.000 0.000
Double seeded
pods./plant 10
0.032 0.407 0.099 0.329 0.508 0.328 0.332 0.653 0.615
0.790 0.000 0.412 0.005 0.000 0.005 0.005 0.000 0.000
Pod weight/plant 11 0.126 0.632 0.198 0.655 0.836 0.276 0.410 0.800 0.909 0.702
0.288 0.000 0.094 0.000 0.000 0.018 0.000 0.000 0.000 0.000
Seed weight 12 0.166 0.648 0.228 0.677 0.840 0.264 0.396 0.788 0.888 0.690 0.990
0.159 0.000 0.053 0.000 0.000 0.024 0.001 0.000 0.000 0.000 0.000
Biomass dry weight 13 0.114 0.636 0.203 0.665 0.854 0.273 0.411 0.807 0.918 0.666 0.987 0.973
0.336 0.000 0.083 0.000 0.000 0.019 0.000 0.000 0.000 0.000 0.000 0.000
Shelling% 14 0.304 -.001 0.294 0.117 0.107 -0.147 -0.165 -0.062 -0.081 -0.074 0.064 0.156 0.025
0.009 0.994 0.011 0.323 0.379 0.216 0.162 0.604 0.497 0.542 0.588 0.186 0.832
100-seed weight 15 0.306 0.445 0.296 0.481 0.456 -0.004 0.176 0.370 0.156 0.234 0.447 0.507 0.418 0.628
0.009 0.000 0.011 0.000 0.000 0.975 0.137 0.001 0.188 0.050 0.000 0.000 0.000 0.000
Page 39 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Table S4: Correlation analysis of recorded traits in the F3 population grown in glasshouses with p-values.
Trait 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Days to emergence 1
Flower
no./plant 2
-
0.158
0.121
leaf no./plant 3
-
0.158
0.212
0.553
0.000
Days to
flowering 4
0.442
0.000
-
0.478
0.000
-
0.391
0.001
Plant height 5
-
0.230
0.068
0.435
0.000
0.410
0.001
-
0.426
0.000
Petiole length 6
-
0.390
0.001
0.442
0.000
0.299
0.017
-
0.477
0.000
0.859
0.000
Leaf area 7
-
0.101
0.426
0.624
0.000
0.873
0.000
-
0.454
0.000
0.456
0.000
0.372
0.002
Terminal leaflet length 8
-
0.128
0.313
0.445
0.000
0.133
0.293
-
0.489
0.000
0.341
0.006
0.379
0.002
0.485
0.000
Terminal
leaflet width 9
-
0.045
0.721
0.344
0.005
0.291
0.020
-
0.308
0.013
0.333
0.007
0.335
0.007
0.645
0.000
0.563
0.000
Plant spread 10
-
0.112
0.379
0.576
0.000
0.385
0.002
-
0.382
0.002
0.254
0.043
0.330
0.008
0.525
0.000
0.449
0.000
0.447
0.000
Stem
no./plant 11
-
0.349
0.005
0.352
0.004
0.443
0.000
-
0.381
0.002
0.583
0.000
0.474
0.000
0.416
0.001
0.212
0.093
0.270
0.031
0.041
0.750
Node
no./stem 12
0.103
0.420
0.568
0.000
0.620
0.000
0.297
0.017
0.142
0.264
0.108
0.394
0.600
0.000
0.257
0.041
0.188
0.138
0.612
0.000
-
0.117
0.358
Internode
length 13
-
0.007
0.958
0.395
0.001
-
0.047
0.713
-
0.321
0.010
0.051
0.691
0.231
0.066
0.139
0.273
0.401
0.001
0.290
0.020
0.722
0.000
-
0.132
0.297
0.343
0.006
Pod no./plant 14
-
0.121
0.342
0.684
0.000
0.763
0.000
-
0.385
0.002
0.448
0.000
0.382
0.002
0.772
0.000
0.331
0.008
0.411
0.001
0.660
0.000
0.262
0.036
0.720
0.000
0.254
0.043
Double
seeded
pods/plant 15
0.219
0.082
0.185
0.143
0.346
0.005
-
0.325
0.009
0.347
0.005
0.379
0.002
0.410
0.001
0.238
0.058
0.309
0.013
0.641
0.000
-
0.057
0.653
0.433
0.000
0.408
0.001
0.554
0.000
Peduncle
length 16
-
0.049
0.703
0.370
0.003
0.065
0.609
-
0.315
0.011
0.035
0.784
0.160
0.208
0.211
0.095
0.394
0.001
0.243
0.053
0.750
0.000
-
0.172
0.173
0.472
0.000
0.799
0.000
0.268
0.032
0.533
0.000
Pod weight/plant 17
-
0.160
0.206
0.604
0.000
0.619
0.000
-
0.418
0.001
0.514
0.000
0.456
0.000
0.728
0.000
0.431
0.000
0.535
0.000
0.701
0.000
0.180
0.156
0.657
0.000
0.287
0.021
0.875
0.000
0.691
0.000
0.414
0.001
Pod length 18
-
0.173
0.171
0.233
0.064
0.134
0.292
-
0.189
0.134
0.463
0.000
0.468
0.000
0.303
0.015
0.493
0.000
0.416
0.001
0.405
0.001
0.291
0.020
0.086
0.500
0.158
0.213
0.171
0.176
0.294
0.018
0.342
0.006
0.438
0.000
Pod width 29
-
0.176
0.164
0.215
0.089
0.121
0.342
-
0.258
0.040
0.465
0.000
0.411
0.001
0.301
0.016
0.466
0.000
0.442
0.000
0.350
0.005
0.327
0.008
0.005
0.968
0.134
0.290
0.106
0.405
0.219
0.083
0.303
0.449
0.387
0.002
0.903
0.000
Pod length of
double seeded 20
-0.178
0.166
0.139
0.281
0.071
0.583
-0.051
0.694
0.361
0.004
0.366
0.003
0.214
0.096
0.356
0.005
0.298
0.018
0.424
0.001
0.104
0.423
0.155
0.299
0.261
0.040
0.217
0.090
0.481
0.000
0.449
0.000
0.453
0.000
0.782
0.000
0.621
0.000
Pod width of
double seeded 21
-
0.032
0.802
0.102
0.428
0.136
0.290
0.033
0.800
0.319
0.011
0.296
0.019
0.325
0.010
0.355
0.005
0.449
0.000
0.320
0.011
0.194
0.131
0.040
0.758
0.033
0.801
0.147
0.253
0.278
0.029
0.264
0.038
0.423
0.001
0.840
0.000
0.839
0.000
0.661
0.000
Seed length 22
-
0.189
0.135
0.223
0.076
0.056
0.660
-
0.237
0.060
0.486
0.000
0.485
0.000
0.261
0.037
0.475
0.000
0.441
0.000
0.237
0.059
0.318
0.011
-
0.052
0.681
0.082
0.520
0.086
0.498
0.222
0.078
0.236
0.061
0.399
0.001
0.803
0.000
0.737
0.000
0.644
0.000
0.723
0.000
Page 40 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome
Draft
Seed width 23
-
0.389
0.001
0.294
0.018
0.051
0.689
-
0.366
0.003
0.432
0.000
0.513
0.000
0.269
0.031
0.536
0.000
0.499
0.000
0.312
0.012
0.326
0.009
0.017
0.894
0.193
0.126
0.114
0.371
0.241
0.055
0.325
0.009
0.423
0.001
0.775
0.000
0.750
0.000
0.601
0.000
0.734
0.000
0.884
0.000
Seed no./plant 24
-
0.145
0.252
0.623
0.000
0.703
0.000
-
0.419
0.001
0.437
0.000
0.405
0.001
0.725
0.000
0.343
0.005
0.416
0.001
0.718
0.000
0.182
0.149
0.725
0.000
0.332
0.007
0.969
0.000
0.698
0.000
0.362
0.003
0.910
0.000
0.187
0.139
0.109
0.391
0.264
0.038
0.153
0.234
0.098
0.441
0.146
0.250
Seed
weight/plant 25
-0.136
0.285
0.645
0.000
0.642
0.000
-0.446
0.000
0.485
0.000
0.438
0.000
0.745
0.000
0.448
0.000
0.523
0.000
0.728
0.000
0.149
0.240
0.713
0.000
0.342
0.006
0.881
0.000
0.680
0.000
0.449
0.000
0.981
0.000
0.382
0.002
0.326
0.008
0.389
0.002
0.342
0.007
0.374
0.002
0.400
0.001
0.920
0.000
Biomass dry
weight 26
-
0.099
0.436
0.671
0.000
0.667
0.000
-
0.414
0.001
0.445
0.000
0.394
0.001
0.774
0.000
0.441
0.000
0.520
0.000
0.754
0.000
0.137
0.279
0.777
0.000
0.371
0.003
0.886
0.000
0.644
0.000
0.486
0.000
0.948
0.000
0.363
0.003
0.310
0.013
0.382
0.002
0.308
0.015
0.298
0.017
0.323
0.009
0.9-3
0.000
0.956
0.000
Shelling % 27
0.052
0.683
0.303
0.015
0.232
0.065
-
0.134
0.291
-
0.112
0.380
0.030
0.812
0.200
0.114
0.064
0.614
-
0.029
0.821
0.184
0.146
-
0.104
0.412
0.333
0.007
0.279
0.025
0.174
0.169
0.049
0.699
0.129
0.309
0.066
0.604
-
0.298
0.017
-
0.334
0.007
-
0.282
0.026
-
0.392
0.002
-
0.104
0.412
-
0.125
0.324
0.190
0.133
0.234
0.063
0.167
0.187
Growth habit 38
0.083
0.512
-
0.245
0.051
0.051
0.689
0.364
0.003
0.003
0.978
-
0.140
0.270
-
0.115
0.366
-
0.427
0.000
-
0.246
0.050
-
0.615
0.000
0.049
0.702
0.299
0.017
-
0.793
0.000
-
0.165
0.191
-0349
0.005
-
0.726
0.000
-
0.216
0.086
-
0.314
0.011
-
0.245
0.051
-
0.405
0.001
0.155
0.299
0.158
0.214
-
0.253
0.044
-
0.233
0.064
-
0.243
0.053
-
0.294
0.018
-
0.100
0.430
100-seed weight 29
-
0.060
0.636
0.349
0.005
0.129
0.311
-
0.256
0.041
0.349
0.005
0.317
0.011
0.351
0.004
0.444
0.000
0.457
0.000
0.287
0.022
0.109
0.390
0.200
0.113
0.175
0.166
0.189
0.135
0.224
0.076
0.305
0.014
0.524
0.000
0.549
0.000
0.564
0.000
0.385
0.002
0.503
0.000
0.800
0.000
0.726
0.000
0.206
0.102
0.550
0.000
0.474
0.000
0.242
0.054
-0.113
0.375
Page 41 of 41
https://mc06.manuscriptcentral.com/genome-pubs
Genome