**5. Challenges and perspectives of exploiting diversity of different gene pools**

The introduction of genetic diversity into elite cotton germplasm is difficult and the breeding process is slow. When breeders use new and exotic germplasm sources, which possess desirable genes for crop trait improvements, large blocks of undesirable genes are also introgressed during the recombination between the two parental lines (linkage drag). This linkage drag has limited the use of such germplasm. Therefore, the utilization of useful genetic diversity of the wild germplasm using traditional breeding efforts is challenging due to: 1) hybridization issues between various cotton genomes, 2) sterility issues of interspecific multi-genome hybrids, 3) segregation distortion, 4) photoperiodic flowering of wild cottons and 5) long timescale (10-12 years of efforts) required for successful introgression and recovering superior quality homozygous genotypes using traditional breeding approaches (Abdurakhmonov, 2007). This underlies necessity for the development of new innovative genomics approaches to support and accelerate the traditional efforts of exploiting the genetic diversity in cotton breeding. Continuing the introduction of genetic diversity into cultivated plants is important for reducing crop vulnerability and improving important traits such as yield, fiber quality traits, and disease and pest resistance of the cotton crop.

The most effective utilization of the genetic diversity of *Gossypium* species further requires (1) characterization of candidate gene(s) underlying the phenotypic and agronomic diversities based on genomic information in other species, (2) estimation of molecular diversity, genetic distances, genealogy and phylogeny of gene pools and germplasm groups, (3) acceleration of linkage mapping and marker-assisted selection, (4) development of efficient cotton transgenomics, and (5) sequencing cotton genome(s) (Abdurakhmonov, 2007). Furthermore, (6) it is very important to characterize and describe the existing cotton germplasm collections for both phenotypic and genomic diversity. Consequently, (7) incorporation of information into electronic web-based cotton databases such as cotton DB (http://cottondb.org), Cotton Portal (http://gossypium.info), and the Cotton Diversity Database (http://cotton.agtec.uga.edu; Gingle et al., 2006) as well as further improvement of data management tools are pivotal to facilitate an effective exploitation of the genetic diversity of cotton in the future. Cotton germplasm exchange (8) among collections and research groups is also an imperative part toward this goal (Abdurakhmonov, 2007).

### **6. Characterization of molecular genetic diversity in** *Gossypium* **genus**

Molecular diversity using protein and DNA marker technologies has extensively been studied for accessions from primary and secondary gene pools. Molecular genetic diversity of tertiary gene pool cotton species is poorly explored using molecular marker technology .

#### **6.1 Molecular diversity within primary gene pool**

#### **6.1.1 Upland germplasm**

As mentioned above, application of modern molecular marker technologies, such as isozymes (Wendel & Percy, 1990; Wendel et al., 1992), random amplified polymorphic DNAs – RAPDs (Multani & Lyyon, 1995; Tatineni et al., 1996; Iqbal et al., 1997; Mahmood et al., 2009; Chaudhary et al., 2010), restricted fragment length polymorphisms – RFLPs (Wendel & Brubaker, 1993), amplified fragment length polymorphisms – AFLPs (Pilay & Myers, 1999; Abdalla et al., 2001; Iqbal et al., 2001; Rana et al., 2005; Lukonge et al., 2007) and Simple Sequence Repeats – SSRs (Liu et al., 2000, Gutierrez et al., 2002; Rungis et al., 2005; Zhang et al., 2005a; Bertini et al., 2006; Zhang et al., 2011a; Kalivas et al., 2011) generally revealed a low level of genetic diversity within Upland cultivars. There were little variations in estimation of molecular diversity among Upland cultivars (*G. hirsutum)*; however, in general, the genetic distance reported for Upland cultivars was in the range of 0.01-0.28 (Abdurakhmonov, 2007).

Recently, we analyzed a large number of *G. hirsutum* variety and exotic accessions from Uzbek cotton germplasm collection (Fig.2) with SSR markers (Abdurakhmonov et al., 2008, 2009). Analysis of a large number of *G. hirsutum* accessions from exotic germplasm and diverse ecotypes/breeding programs with SSR markers confirmed the narrow genetic base of Upland cotton cultivar germplasm pool (with the genetic distance (GD) range of 0.005-0.26) and provided an additional evidence for the occurrence of a genetic 'bottleneck' during domestication events of the Upland cultivars at molecular level (Iqbal et al., 2001). Molecular diversity analysis of germplasm accessions using principal component analysis (PCA) suggested that germplasm resources could be broadly grouped into three large clusters (Fig.3) of exotic (1), USA-type (2) and Uzbekistan (3). First three eigenvalues of PCA analysis accounted for a ~52% variation and demonstrated existence of wide genetic diversity within the exotic germplasm, including germplasm accessions from Mexican and African origin (GD=0.02-0.50; Fig.3). We recorded a plenty of private SSR alleles within each group of accessions, specific to the germplasm groups, breeding ecotypes or exotic accessions.

A wider genetic diversity in the land race stocks of *G. hirsutum* was reported by previous studies (Liu et al., 2000, Lacape et al., 2007), suggesting the existence of sufficient genetic diversity in the exotic germplasm for future breeding programs. Rana et al. (2005) also reported a wider genetic diversity (30-87%) within *G. hirsutum* breeding lines using AFLP markers. Some recent studies have reported a relatively higher genetic diversity with an average genetic distance of up to ~37-77% in *G. hirsutum* cultivars, based on the analysis of specific germplasm resources from Brazil (Bertini et al., 2006), Pakistan (Khan et al., 2009; Azamat & Khan, 2010), India (Chaudhary et al., 2010) and China (Liu et al., 2011; Zhang et al., 2011a) breeding programs. Results of these studies were inferred from SSR or combination of a SSR and/or RAPD marker polymorphisms.

Similarly, using SSR and RAPD markers, Sapkal et al. (2011) reported moderately high level of genetic diversity (up to 57%) for 91 Upland cotton accessions with genetic male sterility maintainer and restorer properties. This suggested the existence of useful genetic diversity both in exotic and breeding line resources, useful to broaden the genetic base of Upland cotton cultivars. There is a need for evaluation of molecular genetic diversity level (Zhang et al., 2011a) and its effective exploitation in breeding programs that will address current concerns on narrowness of genetic base of widely grown Upland cotton cultivars (Hinze et al., 2011).

#### **6.1.2 Sea Island germplasm**

The molecular genetic diversity within *G. barbadense* germplasm accessions was also studied using molecular markers such as allozymes (Wendel & Percy, 1990) and AFLPs (Abdalla et al., 2001; Westengen et al., 2005). These studies revealed a narrow genetic base within *G. barbadense* accessions with a genetic distance of 7-11% (Abdalla et al., 2001; Westengen et al., 2005)

Simple Sequence Repeats – SSRs (Liu et al., 2000, Gutierrez et al., 2002; Rungis et al., 2005; Zhang et al., 2005a; Bertini et al., 2006; Zhang et al., 2011a; Kalivas et al., 2011) generally revealed a low level of genetic diversity within Upland cultivars. There were little variations in estimation of molecular diversity among Upland cultivars (*G. hirsutum)*; however, in general, the genetic distance reported for Upland cultivars was in the range of 0.01-0.28

Recently, we analyzed a large number of *G. hirsutum* variety and exotic accessions from Uzbek cotton germplasm collection (Fig.2) with SSR markers (Abdurakhmonov et al., 2008, 2009). Analysis of a large number of *G. hirsutum* accessions from exotic germplasm and diverse ecotypes/breeding programs with SSR markers confirmed the narrow genetic base of Upland cotton cultivar germplasm pool (with the genetic distance (GD) range of 0.005-0.26) and provided an additional evidence for the occurrence of a genetic 'bottleneck' during domestication events of the Upland cultivars at molecular level (Iqbal et al., 2001). Molecular diversity analysis of germplasm accessions using principal component analysis (PCA) suggested that germplasm resources could be broadly grouped into three large clusters (Fig.3) of exotic (1), USA-type (2) and Uzbekistan (3). First three eigenvalues of PCA analysis accounted for a ~52% variation and demonstrated existence of wide genetic diversity within the exotic germplasm, including germplasm accessions from Mexican and African origin (GD=0.02-0.50; Fig.3). We recorded a plenty of private SSR alleles within each group of accessions, specific to the germplasm groups, breeding

A wider genetic diversity in the land race stocks of *G. hirsutum* was reported by previous studies (Liu et al., 2000, Lacape et al., 2007), suggesting the existence of sufficient genetic diversity in the exotic germplasm for future breeding programs. Rana et al. (2005) also reported a wider genetic diversity (30-87%) within *G. hirsutum* breeding lines using AFLP markers. Some recent studies have reported a relatively higher genetic diversity with an average genetic distance of up to ~37-77% in *G. hirsutum* cultivars, based on the analysis of specific germplasm resources from Brazil (Bertini et al., 2006), Pakistan (Khan et al., 2009; Azamat & Khan, 2010), India (Chaudhary et al., 2010) and China (Liu et al., 2011; Zhang et al., 2011a) breeding programs. Results of these studies were inferred from SSR or

Similarly, using SSR and RAPD markers, Sapkal et al. (2011) reported moderately high level of genetic diversity (up to 57%) for 91 Upland cotton accessions with genetic male sterility maintainer and restorer properties. This suggested the existence of useful genetic diversity both in exotic and breeding line resources, useful to broaden the genetic base of Upland cotton cultivars. There is a need for evaluation of molecular genetic diversity level (Zhang et al., 2011a) and its effective exploitation in breeding programs that will address current concerns on narrowness of genetic base of widely grown Upland cotton cultivars

The molecular genetic diversity within *G. barbadense* germplasm accessions was also studied using molecular markers such as allozymes (Wendel & Percy, 1990) and AFLPs (Abdalla et al., 2001; Westengen et al., 2005). These studies revealed a narrow genetic base within *G. barbadense* accessions with a genetic distance of 7-11% (Abdalla et al., 2001; Westengen et al., 2005)

combination of a SSR and/or RAPD marker polymorphisms.

(Abdurakhmonov, 2007).

ecotypes or exotic accessions.

(Hinze et al., 2011).

**6.1.2 Sea Island germplasm** 

Fig. 3. Principal coordinate analysis of Upland cotton (*G. hirsutum*) accessions from Uzbek cotton germplasm collection analyzed with SSR markers. Two (A) and three (B) dimensional view for accessions from Africa (AF), Afghanistan (AFG), Bulgaria (BUL), China (CN), Czechoslovakia (CZ), India (IN), Korea (KOR), Mexico (MEX), Pakistan (PAK), Syria (SYR), Ukraine (UA), United States (US), Uzbekistan (UZB), Yugoslavia (YU), and others (Turkey, Iraq and Azerbaijan).

as was observed within the Upland cotton germplasm. In contrast, Boopathi et al. (2008) have identified highly diverse pairs of *G. barbadense* accessions using SSR marker analysis, which is useful for breeding of high quality Pima type cotton cultivars. Recently, de Almeida et al. (2009) have studied the molecular diversity level of *G. barbadense* populations *in situ* preserved in the two states of Brazil, Ampa and Para. The genetic analysis using SSR markers of plant populations in these two states revealed 1) high homozygosity in each genotype tested, 2) high total genetic diversity (He=39%) in *G. barbadense* populations studied and 3) high level of population differentiation (Fst=36%) between cotton plants from these two Brazilian states. Results suggested the existence of noticeable genetic diversity preserved in *in situ* populations of *G. barbadense* in Brazil that should be further maintained within an *ex situ* germplasm collection to guarantee its long term preservation (de Almeida et al., 2009). Similarly, there is useful genetic diversity in *ex situ* preserved *G. barbadense* germplasm collections worldwide. For instance, the molecular diversity analysis of *G. barbadense* accessions using SSR markers revealed that moderately higher genetic diversity (up to 34%) exists within former USSR (that includes collections of Uzbekistan and Russia), China, USA, and Egypt germplasm collections (Wu et al., 2010). In that, USSR collection demonstrated the extraordinary genetic diversity compared with other collections whereas Egyptian collection had the least genetic diversity.

#### **6.1.3 Wild allotetraploid germplasm**

The molecular diversity revealed by AFLP markers was low within *G. tomentosum* germplasm with a genetic distance range of 2-11% (Hawkins et al., 2005). However, recent efforts on the characterization of genetic diversity level of three *in situ* preserved *G. mustelinum* population from Brazil using SSR markers suggested 1) high level of homozygosity within each population studied and 2) existence of high level of total genetic differentiations (58.5%) between them, which is due to geographic isolations and genetic founder effects (Barrosso et al., 2010). Wendel & Percy (1990) analyzed 58 *G. darwinii*  accessions from six islands using 17 isozyme markers encoded by 59 genetic loci and identified high genetic diversity level within its accessions and relationships with *G. barbadense* and *G. hirsutum* genomes. This classical study suggested that *G. darwinii* is closely related to *G. barbadense* despite having gene flow imprints from *G. hirsutum;* however, *G. darwinii* has a large number of unique alleles to be considered a distinct genome (Wendel & Percy, 1990).

#### **6.2 Molecular diversity within secondary gene pool**

The genomic diversity of the A-genome diploid cottons has also been studied using molecular marker technology (Liu et al., 2006; Guo et al., 2006; Kebede et al., 2007; Rahman et al., 2008; Kantartzi et al., 2009; Patel et al., 2009; Azamat & Khan, 2010). The genetic distance within 39 *G. arboreum* L (A2A2-genome) accessions, analyzed with SSR markers, ranged from 0.13-0.42 (Liu et al., 2006) demonstrating the existence of wider genomic diversity in the A-genome diploids compared to the Upland cultivar germplasm. Kebede et al. (2007) reported, however, moderate level of genetic diversity within each A1 and A2 genome cottons that ranged from 0.03-0.20 with an average of 0.11 within *G. herbaceum* and 0.02-0.18 with an average of 0.11 for *G. arboreum* (A2). The overall genetic distance between A1 and A2 genomes was up to 36-38% (Kebede et al., 2007; Mahmood et al., 2010). In fact, *G. arboreum* arose from the primitive perennial form of *G. herbaceum* spread in India and there is a single reciprocal chromosomal translocation in *G. arboreum* genome compared to *G. herbaceum* (Guo et al., 2006). Molecular diversity revealed by SSR markers was higher within *G. arboreum* accessions (an average of 25%) compared to *G. herbaceum* accessions (an average of 4%; Patel et al., 2009), suggesting differences in two closely related cotton genome

*in situ* preserved in the two states of Brazil, Ampa and Para. The genetic analysis using SSR markers of plant populations in these two states revealed 1) high homozygosity in each genotype tested, 2) high total genetic diversity (He=39%) in *G. barbadense* populations studied and 3) high level of population differentiation (Fst=36%) between cotton plants from these two Brazilian states. Results suggested the existence of noticeable genetic diversity preserved in *in situ* populations of *G. barbadense* in Brazil that should be further maintained within an *ex situ* germplasm collection to guarantee its long term preservation (de Almeida et al., 2009). Similarly, there is useful genetic diversity in *ex situ* preserved *G. barbadense* germplasm collections worldwide. For instance, the molecular diversity analysis of *G. barbadense* accessions using SSR markers revealed that moderately higher genetic diversity (up to 34%) exists within former USSR (that includes collections of Uzbekistan and Russia), China, USA, and Egypt germplasm collections (Wu et al., 2010). In that, USSR collection demonstrated the extraordinary genetic diversity compared with other collections whereas

The molecular diversity revealed by AFLP markers was low within *G. tomentosum* germplasm with a genetic distance range of 2-11% (Hawkins et al., 2005). However, recent efforts on the characterization of genetic diversity level of three *in situ* preserved *G. mustelinum* population from Brazil using SSR markers suggested 1) high level of homozygosity within each population studied and 2) existence of high level of total genetic differentiations (58.5%) between them, which is due to geographic isolations and genetic founder effects (Barrosso et al., 2010). Wendel & Percy (1990) analyzed 58 *G. darwinii*  accessions from six islands using 17 isozyme markers encoded by 59 genetic loci and identified high genetic diversity level within its accessions and relationships with *G. barbadense* and *G. hirsutum* genomes. This classical study suggested that *G. darwinii* is closely related to *G. barbadense* despite having gene flow imprints from *G. hirsutum;* however, *G. darwinii* has a large number of unique alleles to be considered a distinct genome (Wendel &

The genomic diversity of the A-genome diploid cottons has also been studied using molecular marker technology (Liu et al., 2006; Guo et al., 2006; Kebede et al., 2007; Rahman et al., 2008; Kantartzi et al., 2009; Patel et al., 2009; Azamat & Khan, 2010). The genetic distance within 39 *G. arboreum* L (A2A2-genome) accessions, analyzed with SSR markers, ranged from 0.13-0.42 (Liu et al., 2006) demonstrating the existence of wider genomic diversity in the A-genome diploids compared to the Upland cultivar germplasm. Kebede et al. (2007) reported, however, moderate level of genetic diversity within each A1 and A2 genome cottons that ranged from 0.03-0.20 with an average of 0.11 within *G. herbaceum* and 0.02-0.18 with an average of 0.11 for *G. arboreum* (A2). The overall genetic distance between A1 and A2 genomes was up to 36-38% (Kebede et al., 2007; Mahmood et al., 2010). In fact, *G. arboreum* arose from the primitive perennial form of *G. herbaceum* spread in India and there is a single reciprocal chromosomal translocation in *G. arboreum* genome compared to *G. herbaceum* (Guo et al., 2006). Molecular diversity revealed by SSR markers was higher within *G. arboreum* accessions (an average of 25%) compared to *G. herbaceum* accessions (an average of 4%; Patel et al., 2009), suggesting differences in two closely related cotton genome

Egyptian collection had the least genetic diversity.

**6.2 Molecular diversity within secondary gene pool** 

**6.1.3 Wild allotetraploid germplasm** 

Percy, 1990).

germplasm resources. This is an interesting finding but is in contrast to the report by Kebede et al. (2007) where an average genetic diversity within A1 and A2 genome accessions was equal.

Rahman et al. (2008) studied 32 *G. arboreum* accessions specific to Pakistan with RAPD markers and found up to 53% genetic diversity between studied accessions with very narrow diversity within cultivated *G. arboreum* accessions compared to non-cultivated ones. Analyzing 96 *G. arboreum* accessions with SSR markers, Kantartzi et al. (2009) reported that genetic distance within these geographically diverse A2 genome accessions ranged up to 51%. In a more recent study, Azamat & Khan (2010) also reported wider genetic diversity in *G. arboreum* cultivar germplasm revealed by RAPD (GD=0.371) and SSR markers (GD=0.41). Although variable genetic distance estimates are presented, these reports collectively suggest that A genome representatives of secondary gene pool have sufficient molecular diversity useful for breeding programs.

Studying a large number of accessions for D genome cotton such as *G. aridum* (D4), *G. davidsonii* (D3-d), *G. klotzschianum* (D3-k), G. laxum (D9), G. lobatum (D7), *G. schwendimanii* (D11) with AFLP markers Alvarez & Wendel (2006) have reported 7 to 54% genetic diversity among D-genome accessions studied. A wider range of genetic diversity was observed among 12 D-genome diploid cottons with the genetic similarity of 0.08-0.94 (Guo et al., 2007a), suggesting existence of diverse variations in D-genome cotton germplasm useful for breeding programs. Recently, Feng et al. (2011) have studied 33 arborescent D-genome accessions, including 23 accessions of *G. aridum* with RAPD and AFLP markers. They found high molecular diversity among accessions studied, varying from 32% to 84%. This study suggests for continual efforts to study these D-genome American *Gossypium* species (subsection *Erioxylum*) to resolve genetically distant geographical ecotypes useful for cotton improvement (Feng et al., 2011)

#### **6.3 Molecular diversity within tertiary gene pool**

There is a limited information on molecular diversity estimates for tertiary germplasm pool accessions, including C, E, G and K-genomic species. Recently, Tiwari & Stewart (2008) reported AFLP marker-based molecular diversity analysis results for 57 accessions of C- and G-genome species, including *G. australe* F. Mueller *(G)*, *G. nelsonii* Fryxell (G3), *G. bickii*  Prokhanov (G1) and *G. sturtianum* J.H. Willis (C1). Results showed that within *G. australe* accessions, the pairwise mean genetic distance was in a range of 3-15%, suggesting narrow genetic diversity within *G. australe* accessions that could be due to relatively recent seed dispersal over large growing area of this species (Tiwari & Stewart, 2008). However, there was moderately high molecular diversity between *G. australe* and *G. nelsonii* accessions, ranging from ~17-31%.Higher molecular diversity of up to ~43% was found between *G. australe* and *G. bickii* accessions. The genetic distance between *G. bickii* and *G. nelsonii* varied from 25% to 35% and as expected C1-genome accessions were most distantly related ones to these three G-genome species (Tiwari & Stewart, 2008). There is no report on molecular diversity studies on other representatives of tertiary gene pool.

#### **6.4 Molecular diversity among cotton gene pools**

The genetic diversity among different gene pools was also estimated in many studies using various marker systems. AFLP marker analyses studies (Iqbal et al., 2001, Abdalla et al., 2001, Westengen et al., 2005) revealed that the genetic distance between *G. barbadense* and *G. hirsutum* was in the range of 21-33%. The other wild AD tetraploids (*G. mustelinum, G. tomentosum*) were close to the cultivated AD cottons sharing 75-84% similarity, where *G. tomentosum* was closer to *G. hirsutum* genome (GD=0.16) than the other allotetraploid species (Westengen et al., 2005). At the same time, as mentioned above, *G. darwinii* was closer to *G. barbadense* than *G. hirsutum* (Wendel & Percy, 1990).

Based on AFLP marker analysis, the genetic distance between the widely cultivated AD cottons (*G. barbadense* and *G. hirsutum*) and A-genome diploids varied from 45 to 69%, and that between the cultivated AD cottons and the D-genome varied from 55 to 71%. The genetic distance between the wild AD tetraploids and the A-genome was in the range of 46- 52%, and between the wild AD cottons and the D-genome was 58-59%. The genetic distance between the A- and D-genome cottons was in the range of 0.72-0.82 when analyzed with AFLPs (Iqbal et al., 2001, Abdalla et al., 2001, Westengen et al., 2005).

The use of SSR markers revealed that the genetic distance between *G. hirsutum* and *G. barbadense* was in a range of 42-54% (Kebede et al., 2007). However, Lacape et al. (2007) reported higher genome dissimilarity values (D=0.89-0.91%) between *G. hirsutum* and *G. barbadense* within their material. Also, high mean dissimilarity values were reported between *G. hirsutum* and *G. tomentosum* (D=0.71-0.75) and between *G. barbadense* and *G. tomentosum* (D=0.80) using highly polymorphic sets of SSRs (Lacape et al., 2007). The genetic distance among the AD tetraploids was also in the range of 0.80-0.88 (Liu et al., 2000) with moderate closeness of *G. tomentosum* to the Upland cotton than *G. barbadense* cultivars that was also supported by other studies with different marker systems (Dejoode & Wendel, 1992; Hawkins et al., 2005). Based on SSR marker analysis, the genetic distance between the cultivated AD cottons and the A-genome was in the range of 31-43%, and that between the cultivated AD cottons and the D-genome was in the range of 35-46% (Kebede et al., 2007). The genetic distance between A-and D-genome cottons varied in the range of 29-42% (Kebede et al., 2007).

### **7. Perspectives of 21st century cotton genomics efforts in characterizing and exploiting the genetic diversity of** *Gossypium* **species**

During the past two decades, the international cotton research community has made extensive efforts to utilize the genetic diversity in cotton, which are imperative for the future of trait improvements of the cotton crop. There are many marker systems such as isozymes, RAPDs, RFLPs, AFLPs (extensively referenced herein), and their various modifications (Zhang et al., 2005b) successfully used in cotton. However, the development of a large collection of robust, portable, and PCR-based molecular marker resources such as Simple Sequence Repeats (SSRs; www.cottonmarker.org) and Single Nucleotide Polymorphisms (SNPs) for cotton were one of the tremendous accomplishments of cotton research community (Chen et al., 2007; Van Deynze et al., 2009). This accelerated studies on genetic diversity in cotton at genomic level. Cotton marker resources were made available for cotton research community through cotton marker database (CMD) (Blenda et al., 2006) that are being extensively used to create cotton genetic linkage maps and to map important agronomic QTLs (Abdurakhmonov, 2007; Chen et al., 2007; Zhang et al., 2008). In addition to available DNA marker systems, recently, Reddy et al. (2011) developed a diversity array technology (DArT) marker platform for the cotton genome and evaluated the use of DArT

2001, Westengen et al., 2005) revealed that the genetic distance between *G. barbadense* and *G. hirsutum* was in the range of 21-33%. The other wild AD tetraploids (*G. mustelinum, G. tomentosum*) were close to the cultivated AD cottons sharing 75-84% similarity, where *G. tomentosum* was closer to *G. hirsutum* genome (GD=0.16) than the other allotetraploid species (Westengen et al., 2005). At the same time, as mentioned above, *G. darwinii* was closer to *G.* 

Based on AFLP marker analysis, the genetic distance between the widely cultivated AD cottons (*G. barbadense* and *G. hirsutum*) and A-genome diploids varied from 45 to 69%, and that between the cultivated AD cottons and the D-genome varied from 55 to 71%. The genetic distance between the wild AD tetraploids and the A-genome was in the range of 46- 52%, and between the wild AD cottons and the D-genome was 58-59%. The genetic distance between the A- and D-genome cottons was in the range of 0.72-0.82 when analyzed with

The use of SSR markers revealed that the genetic distance between *G. hirsutum* and *G. barbadense* was in a range of 42-54% (Kebede et al., 2007). However, Lacape et al. (2007) reported higher genome dissimilarity values (D=0.89-0.91%) between *G. hirsutum* and *G. barbadense* within their material. Also, high mean dissimilarity values were reported between *G. hirsutum* and *G. tomentosum* (D=0.71-0.75) and between *G. barbadense* and *G. tomentosum* (D=0.80) using highly polymorphic sets of SSRs (Lacape et al., 2007). The genetic distance among the AD tetraploids was also in the range of 0.80-0.88 (Liu et al., 2000) with moderate closeness of *G. tomentosum* to the Upland cotton than *G. barbadense* cultivars that was also supported by other studies with different marker systems (Dejoode & Wendel, 1992; Hawkins et al., 2005). Based on SSR marker analysis, the genetic distance between the cultivated AD cottons and the A-genome was in the range of 31-43%, and that between the cultivated AD cottons and the D-genome was in the range of 35-46% (Kebede et al., 2007). The genetic distance between A-and D-genome cottons varied in the range of 29-42%

**7. Perspectives of 21st century cotton genomics efforts in characterizing and** 

During the past two decades, the international cotton research community has made extensive efforts to utilize the genetic diversity in cotton, which are imperative for the future of trait improvements of the cotton crop. There are many marker systems such as isozymes, RAPDs, RFLPs, AFLPs (extensively referenced herein), and their various modifications (Zhang et al., 2005b) successfully used in cotton. However, the development of a large collection of robust, portable, and PCR-based molecular marker resources such as Simple Sequence Repeats (SSRs; www.cottonmarker.org) and Single Nucleotide Polymorphisms (SNPs) for cotton were one of the tremendous accomplishments of cotton research community (Chen et al., 2007; Van Deynze et al., 2009). This accelerated studies on genetic diversity in cotton at genomic level. Cotton marker resources were made available for cotton research community through cotton marker database (CMD) (Blenda et al., 2006) that are being extensively used to create cotton genetic linkage maps and to map important agronomic QTLs (Abdurakhmonov, 2007; Chen et al., 2007; Zhang et al., 2008). In addition to available DNA marker systems, recently, Reddy et al. (2011) developed a diversity array technology (DArT) marker platform for the cotton genome and evaluated the use of DArT

*barbadense* than *G. hirsutum* (Wendel & Percy, 1990).

(Kebede et al., 2007).

AFLPs (Iqbal et al., 2001, Abdalla et al., 2001, Westengen et al., 2005).

**exploiting the genetic diversity of** *Gossypium* **species** 

Furthermore, researchers have reported several potential candidate genes of many agronomic traits in cotton. Tremendous efforts were made to study molecular basis of one of the most complex, but important traits – cotton fiber development (Abdurakhmonov, 2007; Chen et al., 2007; Zhang et al., 2008). These efforts, including many more recent reports on the dissection of candidate genes that are specifically expressed in developing fibers are undoubtedly imperative for future exploitation of genetic diversity in cotton fiber traits using transgenomics approaches (Arpat et al., 2004; Ruan et al., 2003; Zhang et al., 2011b).

Despite wide spectra of genetic diversity in *Gossypium* genus and extensive cotton genomics efforts, cotton lags behind other major crops for marker-assisted breeding due to limited polymorphism in the cultivated germplasm. This underlies broadening of cultivar germplasm genetic base through mobilization of useful gene variants from other gene pools into cultivated germplasm. There is a need for application of modern innovative genomics tools such as association mapping to identify genetic causatives of natural variations preserved in cotton germplasm resources and their use in plant breeding. Efforts on turning the gene-tagging efforts from bi-parental crosses to natural population or germplasm collections, and from now classical QTL-mapping approach to modern linkage disequilibrium (LD)-based association study should lead to elucidation of *ex situ* conserved natural genetic diversity of worldwide cotton germplasm resources and its effective utilization. LD refers to a historically reduced (non-equilibrium) level of the recombination of specific alleles at different loci controlling particular genetic variations in a population (Abdurakhmonov & Abdukarimov, 2008). Although novel to cotton research, the association genetics strategy is, in fact, highly applicable to the identification of markers linked to fiber quality and yield through the examination of linkage disequilibrium (LD) of DNA-based markers with fiber quality and yield traits in a large, diverse germplasm collection (Abdurakhmonov et al., 2004, 2008, 2009).

Application of association mapping strategy in gene mapping and germplasm characterization gained wider use in cotton. For example, Kantartzi & Stewart (2008) conducted association analysis for the main fiber traits in 56 *G. arboreum* germplasm accessions introduced from nine regions of Africa, Asia and Europe using 98 SSR markers. Association mapping strategy was also applied for tagging fiber traits in the exotic germplasm derived from multiple crosses among *Gossypium* tetraploid species (Zeng et al., 2009). Both of these studies did not quantify the LD level in the population and used marker-trait associations to tag genetic variations contributing to the trait of interest.

Alternatively, to better assess and exploit a molecular diversity of cotton genus, we conducted molecular genetic analyses in a global set of ~1000 *G. hirsutum* L. accessions, one of the widely grown allotetraploid cotton species, from Uzbek cotton germplasm collection. This global set represented at least 37 cotton growing countries and 8 breeding ecotypes as well as wild landrace stock accessions. The important fiber quality (fiber length and strength, Micronaire, uniformity, reflectance, elongation, etc.) traits were measured in two distinct environments of Uzbekistan and Mexico. This study allowed us to quantify the linkage disequilibrium level in the genome of Upland cotton germplasm and to design an "association mapping" study to find biologically meaningful marker-trait associations for important fiber quality traits that accounts for population confounding effects (Yu et al., 2006; Abdurakhmonov & Abdukarimov, 2008). Several SSR markers associated with major fiber quality traits along with donor accessions were identified and selected for MAS programs (Abdurakhmonov et al., 2008, 2009).

Further, with the specific objective of introducing and enriching the currently-applied traditional breeding approaches with more efficient modern MAS tools in Uzbekistan, we began marker-assisted selection efforts based on our association mapping studies mentioned above. For this purpose, we selected (1) a set of twenty three major (Micronaire, fiber strength and length, and elongation) fiber trait-associated DNA markers as a tool to manipulate the transfer of QTL loci during a genetic hybridization; and (2) thirty-seven (11 wild race stocks and 26 variety accessions from diverse ecotypes) donor cotton genotypes that bear important QTLs for fiber traits. These donor genotypes were crossed with 9 commercial cultivars of Uzbekistan (as recipients) in various combinations with the objective of improving one or more of fiber characteristics of these recipients. These 9 parental recipient genomes were first screened with our DNA-marker panel to compare with 37 donor genotypes. The polymorphic status of marker bands between donor and recipient genotypes were recorded. The hybrid plants generated from each crossing combination were tested using DNA-markers at the seedling stage, and hybrids bearing DNA-marker bands from donor plants were selected for further backcross breeding (Abdurakhmonov et al., 2011).

Testing the major fiber quality traits using HVI in trait-associated marker-band-bearing hybrids revealed that mobilization of the specific marker bands from donors really had positively improved the trait of interest in recipient genotypes (data not shown). Currently, we developed a second generation of recurrent parent backcrossed hybrids (F1BC2), bearing novel marker bands and having superior fiber quality compared to original recipient parent (lacking trait-associated SSR bands). These results showed the functionality of the traitassociated SSR markers detected in our association mapping efforts in diverse set of Upland cotton germplasm. Using these effective molecular markers as a breeding tool, we aim to pyramid major fiber quality traits into single genotype of several commercial Upland cotton cultivars of Uzbekistan. Our efforts will not only help rapid introgression of novel polymorphisms, broadening the genetic diversity of cotton cultivars and accelerating the breeding efforts for future sustainable cotton production in Uzbekistan but also exemplify effective exploitation of the natural genetic diversity *ex situ* preserved in cotton germplasm collections (Abdurakhmonov et al., 2011).

In spite of successful application of association mapping in cotton, there is a great challenge with assigning correct allelic relationships (identity by decent) of multiple band amplicons when diverse, reticulated, and polyploid cotton germplasm resources lacking historical pedigree information are investigated. Besides, there is the issue of rare and unique alleles that is problematic for conducting association mapping (Abdurakhmonov & Abdukarimov, 2008). While these issues can be solved using many available methodologies and approaches (Abdurakhmonov & Abdukarimov, 2008); however, recent studies in model crops suggested a new methodology to minimize these issues with the creation of segregating populations, performing genetic crosses between several reference populations with known allele frequencies for functional polymorphisms. Such an approach is referred to as nested

"association mapping" study to find biologically meaningful marker-trait associations for important fiber quality traits that accounts for population confounding effects (Yu et al., 2006; Abdurakhmonov & Abdukarimov, 2008). Several SSR markers associated with major fiber quality traits along with donor accessions were identified and selected for MAS

Further, with the specific objective of introducing and enriching the currently-applied traditional breeding approaches with more efficient modern MAS tools in Uzbekistan, we began marker-assisted selection efforts based on our association mapping studies mentioned above. For this purpose, we selected (1) a set of twenty three major (Micronaire, fiber strength and length, and elongation) fiber trait-associated DNA markers as a tool to manipulate the transfer of QTL loci during a genetic hybridization; and (2) thirty-seven (11 wild race stocks and 26 variety accessions from diverse ecotypes) donor cotton genotypes that bear important QTLs for fiber traits. These donor genotypes were crossed with 9 commercial cultivars of Uzbekistan (as recipients) in various combinations with the objective of improving one or more of fiber characteristics of these recipients. These 9 parental recipient genomes were first screened with our DNA-marker panel to compare with 37 donor genotypes. The polymorphic status of marker bands between donor and recipient genotypes were recorded. The hybrid plants generated from each crossing combination were tested using DNA-markers at the seedling stage, and hybrids bearing DNA-marker bands from donor plants were selected for further backcross breeding

Testing the major fiber quality traits using HVI in trait-associated marker-band-bearing hybrids revealed that mobilization of the specific marker bands from donors really had positively improved the trait of interest in recipient genotypes (data not shown). Currently, we developed a second generation of recurrent parent backcrossed hybrids (F1BC2), bearing novel marker bands and having superior fiber quality compared to original recipient parent (lacking trait-associated SSR bands). These results showed the functionality of the traitassociated SSR markers detected in our association mapping efforts in diverse set of Upland cotton germplasm. Using these effective molecular markers as a breeding tool, we aim to pyramid major fiber quality traits into single genotype of several commercial Upland cotton cultivars of Uzbekistan. Our efforts will not only help rapid introgression of novel polymorphisms, broadening the genetic diversity of cotton cultivars and accelerating the breeding efforts for future sustainable cotton production in Uzbekistan but also exemplify effective exploitation of the natural genetic diversity *ex situ* preserved in cotton germplasm

In spite of successful application of association mapping in cotton, there is a great challenge with assigning correct allelic relationships (identity by decent) of multiple band amplicons when diverse, reticulated, and polyploid cotton germplasm resources lacking historical pedigree information are investigated. Besides, there is the issue of rare and unique alleles that is problematic for conducting association mapping (Abdurakhmonov & Abdukarimov, 2008). While these issues can be solved using many available methodologies and approaches (Abdurakhmonov & Abdukarimov, 2008); however, recent studies in model crops suggested a new methodology to minimize these issues with the creation of segregating populations, performing genetic crosses between several reference populations with known allele frequencies for functional polymorphisms. Such an approach is referred to as nested

programs (Abdurakhmonov et al., 2008, 2009).

(Abdurakhmonov et al., 2011).

collections (Abdurakhmonov et al., 2011).

association mapping (NAM) and NAM populations would greatly enhance the power of association mapping in plants (Stich and Melchinger, 2010). The usefulness and feasibility of NAM population based genetic mapping studies were successfully demonstrated in maize (Kump et al., 2011; Poland et al., 2011; Tian et al., 2011;) and should be adopted for other crops with complex genome and diverse germplasm resources like cotton. Therefore, creation of NAM populations for cotton on the basis of germplasm evaluation and characterization studies is the task of high priority for future characterization and mapping of biologically meaningful genetic variations in cotton. This requires further efforts and investments that facilitate fine-scale association mapping studies in cotton. This will ultimately lead to cloning and characterization of genetic causatives controlling the genetic diversities and its effective exploitation in plant breeding.

#### **8. Conclusions**

In conclusion, by having a wide geographic and ecological dispersal, the *Gossypium* genus represents and preserves large amplitude of morphobiological and genetic diversity within its *ex situ* worldwide germplasm collections and *in situ* occupation sites. Because of the development of molecular marker technologies, and their application in genetic diversity studies of germplasm resources, various gene pools and specific cultivar groups, researchers found a genetic bottleneck in cultivated cotton germplasm resources. However, there is moderately high molecular diversity present in some specific cultivar germplasm analyzed worldwide, suggesting a need for continual efforts on searching the diverse cultivar germplasm resources using molecular markers. There is a need to extend molecular markerbased diversity studies for tertiary gene pool accessions of cotton. There also is high genetic diversity available within exotic land race stocks, wild AD cottons, and putative A-and Dgenome ancestors to AD cottons that have potential to search for genetic variations useful in future improvement of cotton. The variations observed in genetic distance estimations between different studies could be due to (1) germplasm resources chosen for the study, (2) number of accessions analyzed with molecular markers, (3) number of markers and marker types used, (4) genomic regions screened, and (5) subjective features of data analyses process, e.g., considering or removing unique or rare alleles, largely influencing the genetic distance measures.

Further, the narrowness of genetic diversity in cultivar germplasm was associated with recent and possibly future declines in cotton production and its quality, which was a timely warning to accelerate efforts on broadening the genetic base of cultivar germplasm resource via mobilizing novel genetic variants from wild, primitive, pre-domesticated primary, secondary and tertiary gene pools. Traditional efforts have succeeded in introgressing many new genetic variations into cultivar germplasm from other gene pools, but it is still challenging and the breeding process is slow due to a number of genetic barriers and obstacles, as highlighted above, to accomplish the goal. This underlies the importance of development of innovative tools to exploit the biologically meaningful genetic variations, existing in *Gossypium* genus. The most effective utilization of genetic diversity of *Gossypium* species further requires characterization of candidate gene(s) underlying the phenotypic and agronomic diversities, acceleration of linkage mapping, map-based cloning and markerassisted selection that underlie development of modern genomics technologies such as highresolution, cost effective LD-based association mapping for cotton with its optimization through development of modern nested association mapping populations. The development of efficient cotton transgenomics tools and complete sequencing of cotton genome(s) will further accelerate exploitation of genetic diversity in highly specific manner and with clear vision. Future application of whole genome-association strategy with epigenomics perspectives, which currently is widely being applied in human and the other model plants such as Arabidopsis, will have a significant impact on identifying true functions of genes controlling available genetic diversity, and consequently, its effective utilization.

#### **9. Acknowledgements**

We thank the Office of International Research Programs (OIRP) of United States Department of Agriculture (USDA) for continual funding of our collaborative research on cotton germplasm characterization and genetic diversity analysis. We acknowledge Civilian Research Development Foundation (CRDF), USA for project coordination and Academy of Sciences of Uzbekistan for their continual in-house support of the research efforts.

#### **10. References**


through development of modern nested association mapping populations. The development of efficient cotton transgenomics tools and complete sequencing of cotton genome(s) will further accelerate exploitation of genetic diversity in highly specific manner and with clear vision. Future application of whole genome-association strategy with epigenomics perspectives, which currently is widely being applied in human and the other model plants such as Arabidopsis, will have a significant impact on identifying true functions of genes

We thank the Office of International Research Programs (OIRP) of United States Department of Agriculture (USDA) for continual funding of our collaborative research on cotton germplasm characterization and genetic diversity analysis. We acknowledge Civilian Research Development Foundation (CRDF), USA for project coordination and Academy of

Abdalla, A.M.; Reddy, O.U.K.; El-Zik, K.M. & Pepper, A.E. (2001). Genetic Diversity and

Abdukarimov, A.; Djataev, S. & Abdurakhmonov, I.Y. (2003). Cotton Research in

Abdurakhmonov, I.Y. (2007). Exploiting Genetic Diversity, *Proceedings of World Cotton Research Conference-4*, P2153, Lubbock, Texas, USA, September 10-14, 2007 Abdurakhmonov, I. Y. & Abdukarimov A. (2008). Application of association mapping to

Abdurakhmonov, I.Y.; Abdullaev, A.; Rizaeva, S.; Buriev, Z.; Adylova, A.; Abdukarimov,

Abdurakhmonov, I.Y.; Abdullaev, A.A.; Saha, S.; Buriev, Z.T.; Arslanov, D.; Kuryazov, Z.;

Abdurakhmonov, I.Y.; Buriev, Z.T.; Salakhuddinov, I.B.; Rizaeva, S.M.; Adylova, A.T.

Relationships of Diploid and Tetraploid Cottons Revealed Using AFLP. *Theoretical and Applied Genetics*, Vol. 102, No. 2-3, (February 2001), pp. 222-229, ISSN 1432-2242

Uzbekistan: Elite Varieties and Future of Cotton Breeding, *Proceedings of World Cotton Research Conference-3*, pp. 5-16, Cape Town, South Africa, March 9-13, 2003 Abdullaev, A.A.; Klyat, V.P. & Rizaeva, C.M. (2009). Cotton Introduction in Uzbekistan -

History and Perspectives of Using of Plant Introduction: Problems and Perspectives. (2009), *Proceedings of 4th National scientific-applied conference*, pp. 59-61,

understanding the genetic diversity of plant germplasm resources. *International Journal of Plant Genomics,* Vol. 2008, No. 574927, (June 2008), pp.1-18, ISSN 1687-

A.; Saha, S.; Kohel, R.; Yu, J. & Pepper, A.E. (2004). Evaluation of *G. hirsutum* Exotic Accessions From Uzbek Cotton Germplasm Collection for Further Molecular Mapping Purposes, *Proceedings Beltwide Cotton Improvement Conference,* pp. 1133-

Mavlonov, G.T.; Rizaeva, S.M.; Reddy, U.K.; Jenkins, J.N.; Abdullaev, A. & Abdukarimov, A. (2005). Simple Sequence Repeat Marker Associated With a Natural Leaf Defoliation Trait in Tetraploid Cotton. *Journal of Heredity*, Vol. 96, No.

Shermatov, S.E.; Adukarimov, A.; Kohel, R.J.; Yu, J.Z.; Pepper, A.E .; Saha, S. & Jenkins, J.N. (2006). Characterization of *G. hirsutum* Wild and Variety Accessions

controlling available genetic diversity, and consequently, its effective utilization.

Sciences of Uzbekistan for their continual in-house support of the research efforts.

Tashkent, Uzbekistan, July 3-4, 2009 (in Russian)

1142, San Antonio, Texas, USA, January 5-9, 2004

6, (September 2005), pp. 644-653, ISSN 1465-7333

**9. Acknowledgements** 

**10. References** 

5389

from Uzbek Cotton Germplasm Collection for Morphological and Fiber Quality Traits and Database Development, *Proceedings Beltwide Cotton Improvement Conference,* P5306, San Antonio, Texas, USA, January 3-6, 2006


Boopathi, N.M.; Gopikrishnan, A.; Selvam, N.J.; Ravikesavan, R.; Iyanar, K.; Muthuraman, S.

Brubaker, C.L.; Paterson, A.H. & Wendel, J.F. (1999). Comparative Genetic Mapping of

Campbell, B.T.; Saha, S.; Percy, R.; Frelichowski, J.; Jenkins, J.N.; Park, W.; Mayee, C.D.;

Chaudhary, L.; Sindhu, A.; Kumar, M. & Saini, M. (2010). Estimation of Genetic Divergence

*Physioogyl*, Vol. 145, No. 4, (December 2007), pp. 1303-1310, ISSN 1532-2548 De Almeida, V.C.; Hoffman, L.V.; Yokomizo G.K.I.; da Costa, J.N.; Giband, M. & Barroso,

Dejoode, D.R. & Wendel, J.F. (1992). Genetic Diversity and Origin of the Hawaiian-Islands

Dighe, N.D.; Robinson, A.F.; Bell, A.A.; Menz, M.A.; Cantrell, R.G & Stelly, D.M. (2009).

Endrizzi, J.; Turcotte, E. & Kohel, R. (1985). Genetics, Cytology and Evolution of *Gossypium*.

Esbroeck, G. & Bowman, D. (1998). Cotton Germplasm Diversity and its Importance to

Feng, Ch.; Ulloa, M.; Perez-M, C. & Stewart, J.McD. (2011). Distribution and Molecular

Fryxell, P.A. (1979). Natural History of the Cotton Tribe. Texas A&M University Press,

*Advances in Genetics*, Vol. 23, pp. 271-375, ISBN 9780120176236

No. 7, (July 2009), pp. 719-725, ISSN 0100-204X

(November 1992), pp. 1311-1319, ISSN 0002-9122

(July-August 2009), pp.1151-1164, ISSN 0011-183X

129, ISSN 1523-6919.

2011), pp.615-624, ISSN 1916-2790

College Station, Texas, USA

Crop Science, Vol. 2, No. 3, (March 2010), pp. 039-043, ISSN 2006-9758 Chen, Z.J.; Scheffler, B.E.; Dennis, E.; Triplett, B.A.; Zhang, T.; Guo, W.; Chen, X.; Stelly,

8619

1435-0653

1999), pp. 184-203, ISSN 1480-3321

& Saravanan, N. (2008). Genetic Diversity Assessment of G. barbadense Accessions to Widen Cotton (*Gossypium* spp.) Gene Pool for Improved Fibre Quality. *Journal of Cotton Research and Development*, Vol. 22, No. 2, (July 2008), pp. 135-138, ISSN 0972-

Allotetraploid Cotton and Its Diploid Progenitors. *Genome*, Vol. 42, No. 2, (April

Gotmare, V.; Dessauw, D.; Gband, M.; Du, X.; Jia, Y.; Constable, G.; Dillon, S.; Abdurakhmonov, I.Y.; Abdukarimov, A.; Rizaeva, S.M.; Abdullaev, A.A.; Barrose, P.A.V.; Padua, J.G.; Hoffman, L.V. & Podolnaya, L. (2010). Status of Global Cotton Germplasm Resources. *Crop Science*, Vol. 50, No. 4, (July 2010), pp. 1161-1179, ISSN

Among Some Cotton Varieties by RAPD Analysis. Journal of Plant Breeding and

D.M.; Rabinowicz, P.D.; Town, C.D.; Arioli, T.; Brubaker, C.; Cantrell, R.G.; Lacape, J.M.; Ulloa, M.; Chee, P.; Gingle, A.R.; Haigler, C.H.; Percy, R.; Saha, S.; Wilkins, T.; Wright, R.J.; Van Deynze, A.; Zhu, Y.; Yu, S.; Abdurakhmonov, I.; Katageri, I.; Kumar, P.A.; Mehboob-Ur-Rahman, Z.Y.; Yu, J.Z.; Kohel, R.J.; Wendel, J.F. & Paterson, A.H. (2007). Toward Sequencing Cotton (*Gossypium*) Genomes. *Plant* 

P.A.V. (2009). In situ Genetic Characterization of *Gossypium barbadense* Populations from the States of Para and Amapa, Brazil. *Pesquisa Agropecuaria Brasileira*, Vol. 44,

Cotton, *Gossypium tomentosum*. *American Journal of Botany,* Vol. 79, No. 11,

Linkage Mapping of Resistance to Reniform Nematode in Cotton Following Introgression from *Gossypium longicalyx* (Hutch.& Lee). *Crop Science*, Vol. 49, No. 4,

Cultivar Development. *Journal of Cotton Science*, Vol. 2, No. 3, (July 1998), pp. 121-

Diversity of Arborescent *Gossypium* Species. *Botany*, Vol. 89, No. 9, (September


using Simple Sequence Repeat Molecular Markers. *Genetics and Molecular Research*, Vol. 10, No. 1, (February 2011), pp. 208-217, ISSN 1676-5680


Kantartzi, S.K. & Stewart, J.McD. (2008). Association Analysis of Fibre Traits in *Gossypium* 

Kantartzi, S.K.; Ulloa, M.; Sacks, E. & Stewart, J.M. (2009). Assessing Genetic Diversity in

Kebede, H.; Burow, G.; Dani, R.G. & Allen, R.D. (2007). A-genome Cotton as a Source of

Kump, K.L.; Bradbury, P.J.; Wisser, R.J.; Buckler, E.S.; Belcher, A.R.; Oropeza-Rosas, M.A.;

Lacape, J.M.; Dessauw, D.; Rajab, M.; Noyer, J.L. & Hau, B. (2007). Microsatellite Diversity in

Liu, D.; Guo, X.; Lin, Z.; Nie, Y. & Zhang, X. (2006). Genetic Diversity of Asian Cotton

Liu, S.; Cantrell, R.G.; McCarty, J.C.J. & Stewart, J.M. (2000). Simple Sequence Repeat-based

Liu, Z.; Zhang, Y.; Zhang, S.; Deng, K.; Dong, S. & Ren, Z. (2011). Genetic Diversity Analysis

Luconge, E.; Herselman, L. & Labuschagne, M.T. (2007). Genetic Diversity of Tanzanian

Mahmood, Z.; Raheel, F.; Dasti, A.A.; Shahzadi, S.; Athar, M. & Qayyum, M. (2009). Genetic

Mauer, F. M. (1954). Origin and Taxonomy of Cotton, In: *Cotton*, 383, Academy of Sciences

McCarty, J.C.Jr. & Jenkins, J.N. (1993). Registration of 79 Day-neutral Primitive Cotton

40, No. 5, (September 2000), pp. 1459-1469, ISSN 1435-0653

*Crop Evolution,* Vol. 54, No. 4, (June 2007), pp. 885-895, ISSN 1573-5109 Khan, A.I.; Fu, Y. & Khan, I. (2009). Genetic Diversity of Pakistani Cotton Cultivars as

*Genetics*, Vol. 43, No. 2, (January 2011), pp. 163-168, ISSN 1546-1718

Vol. 10, No. 1, (February 2011), pp. 208-217, ISSN 1676-5680

*Genetica*, Vol. 136, No.1, (May 2009), pp. 141-147, ISSN *1573-6857* 

*Science*, Vol. 4, No. 1, (2009), pp. 21-30, ISSN 1896-0782

1439-0523

ISSN 1572-9788

1573-5109

1684-5315

1435-0653

October 27-31, 2007

of USSR, Tashkent, Uzbekistan (In Russian)

using Simple Sequence Repeat Molecular Markers. *Genetics and Molecular Research*,

*arboreum* Accessions. *Plant Breeding,* Vol. 127, No. 2, (April 2008), pp. 173-179, ISSN

Gossypium arboreum L. Cultivars Using Genomic and EST-derived Microsatellites.

Genetic Variability for Upland Cotton (*Gossypium hirsutum*). *Genetic Resources and* 

Revealed by Simple Sequence Repeat Markers. *Communications in Biometry and Crop* 

Zwonitzer, J.C.; Kresovich, S.; McMullen, M.D.; Ware, D.; Balint-Kurti, P.J. & Holland, J.B. (2011). Genome-Wide Association Study of Quantitative Resistance to Southern Leaf Blight in the Maize Nested Association Mapping Population. *Nature* 

Tetraploid *Gossypium* germplasm: Assembling a Highly Informative Genotyping Set of Cotton SSRs. *Molecular Breeding,* Vol. 19, No. 1, (January 2007), pp. 45-58,

(*Gossypium arboreum L.*) in China Evaluated by Microsatellite Analysis. *Genetic Resources and Crop Evolution,* Vol. 53, No. 6, (September 2006), pp. 1145-1152, ISSN

Assessment of Genetic Diversity in Cotton Race Stock Accessions. *Crop Science*, Vol.

of Various Red Spider Mite-Resistant Upland Cotton Cultivars Based on RAPD. *African Journal of Biotechnology*, Vol. 10, No. 18, (May 2011), pp. 3515-3520, ISSN

Cotton (Gossypium hirsutum L.) Revealed by AFLP Analysis, *Proceedings of African Crop Science Conference*, Vol. 8, pp. 773-776, ISSN 1023-070X, El Minia, Egypt,

Diversity Analysis of the Species of Gossypium by using RAPD Markers. *African Journal of Biotechnology*, Vol. 8, No. 16, (August2009), pp. 3691-3697, ISSN 1684-5315

Germplasm Lines. Crop Science. Vol. 33, No. 2, (March-April 1993), pp. 351, ISSN


Cotton (*Gossypium hirsutum*) from *Gossypium longicalyx*. *Crop Science*, Vol. 47, No. 5, (September-October 2007), pp. 1865-1877, ISSN 0011-183X


Ruan, Y.L.; Llewellyn, D.J. and Furbank, R.T. (2003). Suppression of Sucrose Synthase Gene

Rungis, D.; Llewellyn, D.; Dennis, E.S. & Lyon, B.R. (2005). Simple Sequence Repeat (SSR)

Sapkal, D.R.; Sutar, S.R.; Thakre, P.B.; Patil, B.R.; Paterson, A.H. & Waghmare, V.N. (2011).

Stelly, D.M.; Lacape, J.M.; Dessauw, D.E.G.A.; Kohel, R.; Mergeai, G.; Saha., S.; Sanamyan,

Stich, B. & Melchinger, A.E. (2010). An Introduction to Association Mapping in Plants. *CAB* 

Sunilkumar, G.; Campbell, L.M.; Puckhaber, L.; Stipanovic, R.D. & Rathore K.S. (2006).

Tatineni, V.; Cantrell, R.G. & Davis, D.D. (1996). Genetic Diversity in Elite Cotton

*Nature Genetics*, Vol. 43, No. 2, (January 2011), pp. 159-162, ISSN 1546-1718 Tiwari, R.S. & Stewart, J.McD. (2008). Molecular Diversity and Determination of

Ulloa, M.; Stewart, J.M.; Garcia-C, E.A.; Goday, A. S.; Gaytan-M, A. & Acosta, N.S. (2006).

Ulloa, M.; Percy, R.; Zhang, J.; Hutmacher, R. B.; Wright S. D. & Davis., R. M. (2009).

*Crop Evolution,* Vol. 53, No. 4, (June 2006), pp. 653-668, ISSN 1573-5109 Ulloa, M.; Brubaker, C.; & Chee, P. (2007). Cotton. In: *Genome Mapping & Molecular Breeding,* 

(September-October 2007), pp. 1865-1877, ISSN 0011-183X

Vol.20, No.1, (January-June 2011), pp. 20-28, ISSN 0971-7811

ISBN, Lubbock, Texas, USA, September 10-14, 2007

*Resources*, Vol. 5, No. 039, (May 2010), ISSN 1749-8848

103, No. 48, (November 2006), pp. 18054-18059, ISSN 0027-8424.

Vol. 36, No. 1, (January-February 1996), pp. 186-192, ISSN 1435-0653 Tian, F.; Bradbury, P.J.; Brown, P.J.; Hung, H.; Sun, Q.; Flint-Garcia, S.; Rocheford, T.R.;

2005), pp. 301-307, ISSN 1836-5795

pp. 30-33, ISSN 1942-160X

Heidelberg, Berlin, New York, Tokyo

298X

Cotton (*Gossypium hirsutum*) from *Gossypium longicalyx*. *Crop Science*, Vol. 47, No. 5,

Expression Represses Cotton Fiber Cell Initiation, Elongation, and Seed Development. *The Plant Cell*, Vol. 15, No. 4, (April 2003), pp. 952-964, ISSN 1531-

Markers Reveal Low Levels of Polymorphism Between Cotton (*Gossypium hirsutum L*.) Cultivars. *Australian Journal of Agricultural Research*, Vol. 56, No. 3, (March 23,

Genetic Diversity Analysis of Maintainer and Restorer Accessions in Upland Cotton (*Gossypium hirsutum* L.). *Journal of Plant Biochemistry and Biotechnology*,

M.; Abdurakhmonov, I.Y.; Zhang, T.; Wang, K.; Zhou, B. & Frelichowski, J. (2007). International Genetic, Cytogenetic and Germplasm Resources for Cotton Genomics and Genetic Improvement, *Proceedings World Cotton Research Conference-4*, P. 2154,

*Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition and Natural* 

Engineering Cottonseed for Use in Human Nutrition by Tissue-specific Reduction of Toxic Gossypol. *Proceedings of the National Academy of Sciences of the USA.* Vol.

Germplasm Determined by Morphological Characteristics and RAPD. *Crop Science,*

McMullen, M.D.; Holland, J.B. & Buckler, E.S. (2011). Genome-Wide Association Study of Leaf Architecture in the Maize Nested Association Mapping Population.

Hybridization Among the G-genome *Gossypium* Species, In: *Summaries of Arkansas Cotton Research* 2008, Research Series 573, (October 2008), D.M. Oosterhuis, (Ed.),

Cotton Genetic Resources in the Western States of Mexico: *in situ* Conservation Status and Germplasm Collection for *ex situ* Preservation. *Genetic Resources* and

*Vol. 6: Technical Crops*. C. Kole (ed), pp.1-49. Springer, ISBN 9783540345374,

Registration of four Pima cotton germplasm lines having good levels of *Fusarium*

wilt race 4 resistance with moderate yields, and good fibers. *J. Plant Registr*. Vol. 3, No. 2, (May 2009), pp. 198-202, ISSN 1936-5209

