Microsatellite-based QTL mapping for yield and horticultural traits in cacao (Theobroma cacao L.)

 Corresponding author: Joseph Moroh AKAZA Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution License 4.0. Microsatellite-based QTL mapping for yield and horticultural traits in cacao (Theobroma cacao L.) Joseph Moroh AKAZA 1, *, Bakari Abou KOUASSI 2, Koffié KOUAKOU 1, Olivier FOUET 3, Simon-Pièrre Assanvo N’GUETTA 2, Sélastique Doffou AKAFFOU 1 and Claire LANAUD 3 1 Agricultural production improvement Laboratory, Department of Biology Physiology Genetics, Faculty of Agroforestry, Jean LOROUGNON GUEDE University, Po Box 150 Daloa, Côte d’Ivoire. 2Félix Houphouët-Boigny University, Faculty of Biosciences, Laboratory of Biotechnologies, Agriculture and biological resources utilisation, Genetics pedagogical and research Unit, 22 Po. Box 582, Abidjan 22, Côte d’Ivoire. 3 CIRAD-Bios, UMR AGAP Institute, Avenue Agropolis, 34398 Montpellier Cedex 5, France. GSC Advanced Research and Reviews, 2025, 25(02), 390-406 Publication history: Received on 02 October 2025; revised on 22 November 2025; accepted on 25 November 2025 Article DOI: https://doi.org/10.30574/gscarr.2025.25.2.0351 Abstract In cacao, a perennial crop, SSR could ease/facilitate and speed up/accelerate selection for yield, vigour and their components. This investigation was conducted to identify Quantitative Trait Loci (QTL) underlying these traits in three hybrid progenies created, at CNRA in Côte d'Ivoire, from crosses (SCA6 x H) x C1, (P7 x ICS100) x C1 and (P7 x ICS95 x) x C1 and, encompassing, respectively, 179, 173 and 183 plants. Yield related-traits were recorded during three full campains years. In each progeny, Kruskal-Wallis test and interval mapping were applied on a SSR based-linkage map. QTLs detected are: 18 for yield components (healthy ripe pods number, pods weight, dried beans weight, one-hundred dried beans weight) on chromosomes 1, 2, 4, 6, 10, with a major for PW (R2 = 34.3 %) borded by markers mTcCIR168 and mTcCIR18 on chromosome 4 of the parent derived from “P7 x ICS95”, two for trunk girth on chromosome 4, three for yield efficiency. Some clusters of QTLs of yield components were observed on chromosomes 1, 4 and 6, 6 suggesting pleiotropic and/or polysgenic effects. The chromosome 4 carried most of QTLs. Some QTLs of the healthy ripe pods number, pods weight, dried beans weight, stable across progenies, genetic backgrounds and years, and moreover linked to markers mTcCIR168 and smTcCIR18, are useful in breeding programs. Keywords: Theobroma cacao L.; Simple Sequence Repeats (SSR); Quantitative trait loci (QTL); Yield components; Horticultural vigour 1. Introduction Cacao (Theobroma cacao L.) is an important export commodity crop in many countries in Latin America, sub-saharan Africa and Asia. In 2014, [1] reported that cocoa provided a livelihood for over 120 million people worldwide. However, cocoa production has been constrained by a number of factors, including climate changes, low productivity. The annual average production of cocoa beans is very low on farms, particularly in Africa, varying between 218,64 and 600 kilograms per hectare [2, 3]. This low productivity is, to a certain extent, due to genetic facors. Yield and vigour are complex traits controlled by many genes. Their oligogenic inheritance system has been demonstrated and is still being well unknown. Since many decades, steps towards high yield in cocoa breeding programs by means of traditional methods, crosses and generations of phenotypic backcrossing or selfing, have been limited. This phenotypic-based selection is laborious and GSC Advanced Research and Reviews, 2025, 25(02), 390-406 391 time consuming for a slow-growing plant, cacao; in addition to be expensive. Thus, despite these efforts, up to now, no genotype with high yield level has been found or created. Thus, the introduction of molecular markers in the selection process, via markers-assisted selection, should improve the efficiency and should be a convenient alternative to phenotypic selection. Indeed, DNA markers are very efficient and powerful tools in plant breeding for the genetic dissection of quantitative traits and the early screening of desired genotypes in perennial crops, such as cacao, thereby offering new opportunities for genetic improvement in cacao. In cacao, few studies have been conducted to analyze markers-trait associations and to detect QTL for yield and vigor [4, 5, 6, 7, 8, 9, 10, 11]. In these studies, markers were combined: either, markers other than microsatellite or a few SSR with markers of other types. In these cases where different types of markers were mixed, problems occurred in interpreting the results with complex profiles [12] and accordingly should be so in implementing MAS. Microsatellite, (SSR), is an ideal PCR-based DNA marker for genetic mapping and MAS because of their multiple advantageous features [13, 14, 15, 16] that make them more suited for molecular genetic studies in developing countries like Côte d’Ivoire. The present study aimed at identifying SSR markers-based QTLs for yield and its components (pods number, pods weight, beans weight) and adult tree vigor in three related hybrid populations monitored over a 3-year period. QTL stability across time was also analyzed. 2. Materials and methods 2.1. Plant materials The study was conducted on three mapping hybrid progenies, comprising, respectively, 179, 173 and 183 plants, derived from crosses (SCA6 x H) x C1, (P7 x ICS100) x C1 and (P7 x ICS95) x C1. These were set up at the Centre National de Recherche Agronomique (CNRA) research station in Bingerville, South-Eastern Côte d’Ivoire. Their male parent, C1 (or IFC1), is a highly homozygous [17] Lower Amazon Forastero clone local type Amelonado. It is a relatively high or good yielding clone [18, 19] with a relatively good adult vigor [18]. ICS100, ICS95 and H are Trinitario clones, from Trinidad, known for the big size of their beans. ICS95 is one of the best and precious clones owing to its high yielding level and excellent or good pod filling [20, 19, 21]. While ICS100 presents a relatively good yield level and a good adult vigor (trunk girth) [22, 23]. P7 (or Pound 7), an Upper Amazon Forastero clone, from wild origin, was one of the progenitors of commercial varieties in Côte d’Ivoire, especially in crosses with Lower Amazon Forastero clones (Amelonado). It exhibits a good production [23, 21] and a very good adult vigor (trunk girth) [23]. SCA6, an Upper Forastero amazon clone from wild origin [22], is known for, on one hand, and, one the other hand, a high cocoa yielding [22], high yield efficiency and yield/vigour relationship [22, 19]. Thus, C1, P7 and SCA6 are used as parents for high-yielding hybrid families with good pod filling and good grain formation [24, 25, 26, 27, 28]. 2.2. Methods 2.2.1. Experimental design and trial monitoring The three studued hybrid progenies were planted according to a completely randomized mating design of rows with border (guards). Each plant (tree) represented a unique genotype. The plants spacing was 3 m both between rows and in row. Pollination was not controlled. Weedings in the fields were performed manually and, if needed, supplemented with chemical treatments by spraying Kalach 360 SL (360 g/L) at the rate of 10 mL/L water. Pests were controlled with direct applications of Thiodan 50 CE at the rate of 12.5 mL/L water. Regular pruning was performed on the exceeding plagiotropic and orthotropic branches, along with the removal of parasitic epiphytes (Loranthus spp). No fertilizer was applied. 2.2.2. Traits measurements Traits measurements were done on an individual tree basis. GSC Advanced Research and Reviews, 2025, 25(02), 390-406 392 Yield related-traits scoring The following traits were recorded for each of the 179, 173 and 183 trees of the three studied progenies. During four cocoa campaigns (from september to august), cacao healthy ripe pods monthly harvested were weighted. The cocoa yield was expressed as cumulated pods number (PN), pods weight (PW) and dried beans weight (DBW) by transforming the wet beans weight (WBW) as follows: WBW = PW x α, with α a transformation coefficient [20]. DBW = WBW x d x α’, where d is the planting density (in our case d = 1,337 trees/ha) and α’ is a coefficient of transformation of WBW to DBW, generally fixed at 0.35. In addition, for only trees of progenies derived from (P7 x ICS100) x C1 and (P7 x ICS95) x C1, a sample of about 400 g of fresh beans was micro-fermented and sun dried at every harvest. Then, one-hundred dried beans weight (W100) was recorded and its average values were calculated per year and over the three-year period. The yield efficiency (YE) was, also, estimated as the ratio of the cumulated healthy ripe pods weight data over the three years to the trunk cross-section (cm2). Horticultural traits measurement Three horticultural vigor traits, namely the trunk girth (TG), the jorquette height (JH) and the canopy size (CS), were scored in both half-sib families (progenies) derived from crosses (P7 x ICS100) x C1 and (P7 x ICS95) x C1. The trunk girth (TG) was measured at about 10 cm above the soil, in the 6th and 10th years after planting. When the tree presented two trunks, TG has been adjusted as follows: TG = ( ) 2211 TGTGTGTG + [22], TG1 and TG2 being girths of trunks 1 and 2, respectively. The jorquette height (JH) was measured as the height (cm), from the soil surface, to the first jorquette (first point at which trunk breaks into lateral branches), in the 6th and 10th years after planting. When the tree presented two trunks, the highest height value has been taken into account in analyses. The canopy surface (CS) was estimated, in the 10th year after planting, as an approximation of the surface of the crown calculated from its projection on the soil, as follows: CS (m2) = π x CD1/2 x CD2/2, with CD1 the maximal canopy diameter measured in the direction of the row in which the tree is and CD2 the maximal one measured in the perpendicular direction. 2.2.3. Molecular evaluations DNA evaluation and SSR analyses are fully described in [29]. 2.2.4. Data analyses Phenotypic data Normal distribution of the data of each trait was verified applying the chi square test (χ2) and/or the Agostino test at the threshold α = 5 %. Variances homogeneity was also checked by the test of Bartlett and/or the test of Brown-Forsythe. Descriptive statistical analyses (means and coefficients of variation), correlation coefficients (r) according to Pearson among traits and analyses of variance were calculated or performed with STATISTICA 7.1 software (StatSoft Group Inc., Tulsa, Okla.). Trees that did not produce any pod were discarded from descriptive statistical analyses. Natural logarithm transformation, square root transformation or arc sine transformation were applied, according to the type of trait analyzed, in case of evident anormality. Non parametric tests were performed when transformations did not achieve a sufficiently normal distribution. 2.3. Molecular data The linkage maps used for the QTL mapping were constructed and published by [29]. Also, QTL mapping strategy is the one described in this previous paper. 2.3.1. QTL stability across time For analyzing QTL stability across harvesting years, five times were defined: the three full years of the study (monitoring and scoring traits), the main and accessory harvest periods. The main harvest period (from september to january) was defined as the sum of the main harvest periods of the three years and the accessory harvest period (from april to july) as the sum of the accessory harvest periods of these three years. The QTL stability analysis was performed with phenotypic data collected during each of these five times. Thus, a stable QTL (QTLST) is defined as a QTL detected in the three years. While a specific QTL (QTLSP) was a QTL detected only in any one given of the five times (year or period) and in the three progenies simultaneously. GSC Advanced Research and Reviews, 2025, 25(02), 390-406 393 2.3.2. Pleiotropic and polygenic QTLs determination QTLs positions on linkage groups of the different parents were compared and discussed for determining pleiotropic and polygenic QTLs in relation to traits controlled. Pleiotropic QTLs, associated with different traits are, normally, detected in the same genomic region, and polygenic QTLs in different genomic regions. 3. Results 3.1. Phenotypic data distribution Distributions of phenotypic data were normal (p > 0.05) for jorquette height (JH), canopy size (CS), one-hundred dried weight (W100) in both half-sib families, besides for trunk girth (TG) in the family derived from (P7 x ICS95) x C1. But almost normal for TG in the family derived from (P7 x ICS100) x C1 and for TG and JH in the family derived from (P7 x ICS95) x C1. Whereas, a group presented high deviation to normal distribution. This group is composed of healthy ripe pods number (PN), pods weight (PW), dried beans weight (DBW) in the three families, besides yield efficiency (YE) in the family derived from (P7 x ICS100) x C1 and YE in the family derived from (P7 x ICS95) x C1. For these traits, distributions remained non normal despite transformations performed. Non-parametric tests were, thus, applied. Homogeneity of variances was observed for the quasi-totality of traits analyzed. 3.2. Correlations between traits Of the 28 correlations tested among the height traits studied (five for yield: PN, PW, DBW, W100, YE; three for vigor: TG, JH, CS) in each one of the two half-sib progenies, respectively, 17 and 18 were significant (p < 0.05, p < 0.01 and p < 0.001) (Tables 1 and 2). All these correlations are positive. No correlation was found between W100 and the other seven traits. TG and CS have the highest number (six) of significant correlations with the other six traits. The strongest correlations coefficients, ranging from 0.96 to 1.00, were observed, unsurprisingly, among PN, PW and DBW. These yield components (PN, PW, DBW) are strongly correlated (0.77*** ≤ r ≤ 0.81***) with TG and relatively strongly correlated (0.57*** ≤ r ≤ 0.68***) with CS. The JH was correlated only with the other two vigor traits (TG and CS) in the progeny derived from (P7 x ICS100) x C1, while it was correlated with YE, besides these other two vigor traits in the progeny derived from (P7 x ICS95) x C1. In the population derived from (SCA6 x H) x C1, each of the yield traits PN, PW and DBW were highly strongly (0.92 ≤ r ≤ 1.00) correlated with the two others (Table 3). Table 1 Pearson’s phenotypic correlation coefficients among traits analyzed in the population derived from (P7 x ICS100) x C1 PN PW DBW W100 TG JH CS PW 0.96*** DBW 0.96*** 0.99*** W100 Ns Ns Ns TG 0.78*** 0.80*** 0.81*** ns JH Ns Ns Ns ns 0.36*** CS 0.63*** 0.67*** 0.68*** ns 0.81*** 0.23** YE 0.87*** 0.89*** 0.88*** ns 0.64*** Ns 0.55*** ns : non significant ; significant * at p < 5 % ; ** at p < 1 % ; *** at p < 0,1 %.. PN: healthy ripe pods number; PW: pods weight; DBW: dried beans weight; W100: one hundred dried beans weight; TG: trunk girth; JH: jorquette height; CS: canopy size; YE: yield efficiency. Table 2 Pearson’s phenotypic correlation coefficients among traits analyzed in the population derived from (P7 x ICS95) x C1 PN PW DBW W100 TG JH CS PW 0.97*** DBW 0.97*** 1.00*** GSC Advanced Research and Reviews, 2025, 25(02), 390-406 394 W100 Ns Ns Ns TG 0.77*** 0.81*** 0.81** ns JH Ns Ns Ns ns 0.31*** CS 0.59*** 0.58*** 0.57*** ns 0.84*** 0.18* YE 0.94*** 0.93*** 0.93*** ns 0.76*** 0.16* 0.60*** ns : non significant ; significant * at p < 5 % ; ** at p < 1 % ; *** at p < 0,1 %.. PN: healthy ripe pods number; PW: pods weight; DBW: dried beans weight; W100: one hundred dried beans weight; TG: trunk girth; JH: jorquette height; CS: canopy size; YE: yield efficiency. Table 3 Pearson’s phenotypic correlation coefficients among traits analyzed in the population derived from (SCA3 x H) x C1 PN PW PW 0.93*** DBW 0.92*** 1.00*** ns : non significant ; significant; *** at p < 0.1 %; PN: healthy ripe pods number; PW: pods weight; DBW: dried beans weight. 3.3. Plant influence on traits variability The analysis of variance revealed different influences of the factor genotype (a tree = a single genotype) on the variation of traits: significant (p < 5 %) for TG, YE, highly significant (p < 1 %) for PN and DBW, very highly significant (p < 0.1 %) for PW. 3.4. QTL identified QTLs detected in the three progenies for the eight traits anslyzed are summarized in table 4. For each yield related-trait, three to five QTLs were identified on three to five linkage groups. While two QTLs of one horticultural trait was detected on one linkage group. 3.4.1. QTL of yield components sNo QTL of healthy ripe pods number (PN) was found in the parent derived from “P7 x ICS100”. Three QTLs for PN were identified on linkage groups (LGs) 1, 2 in parent derived from “SCA6 x H” and 4 of parent derived from “P7 x ICS100” (Figure 1; Table 4). The most significant were those on LGs 2 and 4 (Table 4). Five QTL of healthy ripe pods weight (PW) were identified (Figure 1; Table 4): two in each parent derived from “SCA6 x H” and “P7 x ICS100” and one in parent derived from “P7 x ICS95”. The LG 4 carried one in two parents. That of the parent derived from (P7 x ICS95) had a major effect ‘(R2 = 34,3 %). No QTL of dried beans weight (DBW) was found in the parent derived from “P7 x ICS95”. Five QTLs for DBW were detected, with two in the parent derived from “SCA6 x H” and three in the parent derived from “P7 x ICS100”. Five QTLs for one-hundred dried beans weight (W100) were identified, with four in the parent derived from ‘’P7 x ICS100” and one in the parent derived from “P7 x ICS95”. The phenotypic variation for the trait explained by these QTLs varied from 15.6 to 20.9 %. Three QTLs of yield efficiency (YE) were mapped: one and two, respectively, in the parent derived from “P7 x ICS100” and “P7 x ICS95”. The LG 6 carried one in both parents. The R2 varied from 8.5 to 13.6 %. 3.4.2. QTL related to plant vigor No QTL was detected for the jorquette height (JH) and the canopy size (CS). Two QTLs of the trunk girth (TG) were found on LG 4 of the parent derived from “P7 x ICS100”, with R2 varying from 10.7 to 11.9 %. GSC Advanced Research and Reviews, 2025, 25(02), 390-406 395 GSC Advanced Research and Reviews, 2025, 25(02), 390-406 396 GSC Advanced Research and Reviews, 2025, 25(02), 390-406 397 GSC Advanced Research and Reviews, 2025, 25(02), 390-406 398 Figure 1 Graphical representation of the linkage groups (LGs) of the three female parents carrying their respective detected QTLs Vertical lanes representing LGs of the female parent “P7 x ICS100” in white, those of the female parent “P7 x ICS95” in white-black and of the female parent “SCA6 x H” in black. On the right side of the LG, the triangle is proportional to the percentage of the phenotypic variance of the trait explained by the QTL. On the left, cumulative map distances in cM and the locus names are indicated. Markers loci with asterisks deviated significantly from a 1:1 ratio. Table 4 Summary of QTL data for yield components, vigour and resistance traits to Phytophthora palmivora Traits Progeny derived from cross chr KW (SMA) SIM K* SL Position Bording marlers LOD peak R2 GSC Advanced Research and Reviews, 2025, 25(02), 390-406 405 [29] Akaza MJ, Kouassi AB, Akaffou DS, Fouet O, N’guetta AS-P, Lanaud C. Mapping QTLs for Black pod (Phytophthora palmivora) resistance in three hybrid progenies of cocoa (Theobroma cacao L.) using SSR markers. International Journal of Scientific and Research Publications. 2016; 6(1): 298-311. [30] Charlson DV, Bhatnagar SCAK, Ray JD, Sneller CH, Carter TEJ, Purcell LC. Polygenic inheritance of canopy wilting in soybean [Glycine max (L.) Merr.]. Theor Appl Genet. 2009; 119:587–594. [31] Zhang X, Han S, Tang F, Xu J, Liu H, Yan M, Dong W, Huang B, Zhu S. Genetic analysis of yield in peanut (Arachis hypogaea L.) using mixed model of major gene plus polygene. African Journal of Biotechnology. 2011; 10(37): 7126-7130. [32] Simpson JL. Polygenic or Multifactorial Inheritance. The Global Library of Women’s YZMedicine [internet]. Stockholm: Kristina Gemzell Danielsson Karolinska University Hospital, Stockholm, Sweden; @ 2012 [cited 2025 october 18]. Available from https://www.glowm.com/sectionview/heading/Polygenic%20or%20Multifactorial%20Inheritance/item/343 [33] Croucha JMD, Bodmerb FW. Polygenic inheritance, GWAS, polygenic risk scores, and the search for functional variants. PNAS. 2020; 117 (32): 18924 – 18933. [34] Cazzola F, Bermejo CJ, Cointry E. Transgressive segregations in two pea F2 populations and their respective F2:3 families. Pesquisa Agropecuária Brasileira. 2020; v.55, e01623. [35] Pramanik K, Kumari M, Sahu GS, Acharya GC, Tripathy p, Dash M, Jena C. Assessment of Transgressive Segregants for Yield and its Component Traits in French Bean (Phaseolus vulgaris L.). Legume Research; 2024. 47(5): 695704. [36] Athira S, and Lovely B. Beyond Parental Limits: The Role of Transgressive Segregation in Modern Plant Breeding. Journal of Advances in Biology & Biotechnology. 2025; 28(9): 498-515. [37] Crouzillat D, Lerceteau E, Petiard V, Morera J, Rodriguez H, Walker D, Phillips W, Ronning C, Schnell R, Osei J, Fritz P. Theobroma cacao L. : a genetic map and quantitative trait loci analysis. Theor Appl Genet 1996; 93:205-214. [38] Sié RS, N’Goran JAK, Montagnon C, Akaffou, DS, Cilas, C, Dagou S, Mondeil F, Charles G, Branchard M. Characterization and evaluation of two genetic groups and value of Intergroup hybrids of Cola nitida (Vent.) Schott and Endlicher. Euphytica. 2009: 167:107-112. [39] Sié RS. Study of the diversity and genetic parameters of cola tree Cola nitida (Vent.) Schott et Endlicher of the collection of Côte d’Ivoire. [PhD Thesis]. University of Cocody (Côte d’Ivoire) ; 1999. [40] Marcano M, Morales S, Hoyer MT, Courtois B, Risterucci AM, Fouet O, Pugh T, Cros E, Gonzalez V, Dagert M, Lanaud C. A genomewide admixture mapping study for yield factors and morphological traits in a cultivated cocoa (Theobroma cacao L.) population. Tree Genetics & Genomes. 2009; 5 : 329–337. [41] N’Goran JAK, Risterucci AM, Clément D., Sounigo O, Lorieux M, Lanaud C. Identification of quantitative trait loci (QTL) in Theobroma cacao L. Agronomie Africaine. 1997; IX (1). [42] Lanaud C, Boult E, Clapperton J, Koffi N, N’Goran J, Cros E, Chapelin M, Risterucci AM, Allaway D, Gilmour M, Cattaruzza A, Fouet O, Clément D, Petithuguenin P. Identification of QTLs related to fat content, seed size and sensorial traits in Theobroma cacao L. In: Cocoa Producers' Alliance, eds. 14th International Cocoa Research Conference. Proceedings. Lagos :2005. p: 1119-1126. [43] Crouzillat D, Rigoreau M, Cabigliera M, Alvarez M, Bucheli P, Pétiard V. QTL Studies Carried Out for Agronomic, Technological and Quality Traits of Cocoa in Ecuador. In: Bekele LF. Proceedings of the International Workshop on New Technologies and Cocoa Breeding, INGENIC, 16th-17th October 2000, Kota Kinabalu, Sabah, Malaysia. London, UK. INGENIC, 2001. p: 120-126. [44] Clément D. Mapping QTL controlling characters of interest in cacao (Theobroma cacao L.). [PhD Thesis]. Agronomic National Institute Paris-Grignon ; 2001. [45] Lv J, Liu N, Guo J, Xu Z, Li X., Li Z., Luo H, Ren X, Huang L, Zhou X, Chen Y, Chen W, Lei Y, Tu J, Jiang H, Liao B. Stable QTLs for Plant Height on Chromosome A09 Identified From Two Mapping Populations in Peanut (Arachis hypogaea L.). Frontiers Plant Science. 2018; 9 (684): 1 – 12. [46] Wang J, Liu H, Zhao C, Tang H, Mu Y, Xu Q, Deng M, Jiang Q, Chen G, Qi P, Wang J, Jiang Y. Chen S, Wei Y, Zheng Y, Lan X, Ma J. Mapping and validation of major and stable QTL for flag leaf size from tetraploid wheat. The Plant Genome. 2022; 15: 1 – 17. GSC Advanced Research and Reviews, 2025, 25(02), 390-406 406 [47] Haugrud ARP, Zhang Q, Green AJ, Xu S, Faris JD. Identification of stable QTL controlling multiple yield components in a durum 3 cultivated emmer wheat population under field and greenhouse conditions. Genes Genomes Genetics. 2023; 13(2) : 1 – 13. [48] Zhang SZ, Hu X-H, Wang F-F, Chu Y, Yang W-G, Xu S, Wang S, Wu L-R, Yu H-L, Miao H-R, Fu C, Chen J. A stable and major QTL region on chromosome 2 conditions pod shape in cultivated peanut (Arachis hyopgaea L.). Journal of Integrative Agriculture. 2023; 22(8): 2323–2334. [49] Lanaud C, Flament MH., Nyassé S, Risterucci AM, Fargeas D, Kébé I, Motilal L, Thévenin J-M, Paulin D, Ducamp M, Clément D, N’Goran JAK, Cilas C. Synthesis of studies on genetic basis of cocoa resistance to Phytophthora using molecular markers. In : Cocoa Producers’ Alliance. Proceedings pf the 13th International Cocoa Research Conference: Towards the effective and optimum promotion of cocoa through research and development. Cocoa Producers’ Alliance, Lagos, Nigeria. 2001. p: 127-135. [50] Santos FFJ, Lopes UV, Pires JL, Melo GRP, Gramacho KP, Clément D. QTLs detection under natural infection of Moniliophtora perniciosa in a cacao F2 progeny with scavina-6 descendants. Agrotrópica. 2014; 26(1): 65 - 72. [51] Mournet P, Beviláqua de Albuquerque PS, Alves RM, Silva-Werneck JO, Rivallan R, Lucilia HM, Clément D. A reference high-density genetic map of Theobroma grandiflorum (Willd. ex Spreng) and QTL detection for resistance to witches’ broom disease (Moniliophthora perniciosa). Tree Genetics & Genomes. 2020. 16 (89) : 113. [52] Fernandes LDS, Royaert S, Corrêa FM, Mustiga GM, Marelli J-P, Corrêa R. X, Motamayor JC. Mapping of a Major QTL for Ceratocystis Wilt Disease in an F1 Population of Theobroma cacao. Frontiers in Plant Science. 2018; 9(155): 1-15. [53] Araújo IS, de Souza Filho GA, Pereira MG, Faleiro FG, de Queiroz VT, Guimarães CT, Moreira MA, de Barros EG, Machado RCR, Pires JL, Schnell R., Lopes UV. Mapping of Quantitative Trait Loci for Butter Content and Hardness in Cocoa Beans (Theobroma cacao L.). Plant Mol Biol Rep. 2009; 27:177–183. [54] Barchi L, Lefebvre V, Sage-Palloix A-M, Lanteri S, Palloix. QTL analysis of plant development and fruit traits in pepper and performance of selective phenotyping. Theor Appl Genet. 2009; 118:1157–1171. [55] Vinarao R, Proud C, Zhang X, Snell P, Fukai S, Mitchell J. Stable and Novel Quantitative Trait Loci (QTL) Confer Narrow Root Cone Angle in an Aerobic Rice (Oryza sativa L.) Production System. Rice. 2021. 14 (28) : 1 – 12.