Assessing Genetic Relatedness of In Vitro Cassava Accessions Using Quantitative Morphological Traits and DarTseq Single Nucleotide Polymorphism (SNP) Markers

doi:10.21203/rs.3.rs-5047468/v1

Genetic diversity is crucial for breeding progress and biodiversity. Genetic relatedness was assessed using quantitative morphological traits (plant height, root length, number of roots, number of leaves, leaf length, leaf width, and number of leaf lobes) and SNP markers of 101 in vitro cassava (Manihot esculenta Crantz) accessions from the International Institute of Tropical Agriculture (IITA), Genetic Resources Centre, in 2023. Analysis of Variance (ANOVA) of morphological traits revealed significant differences (P < 0.01) of all traits across accessions and weeks. Root length and number of leaf lobes had significant differences (P < 0.01) across accessions and weeks. Cluster analysis identified four distinct clusters. Principal Component Analysis (PCA) revealed that the first three components explained 67.26% of the total variation among accessions. The leaf length (LL), Leaf width (LW), Number of roots (NR), Plant height (PH), and Root Length (RL) had the highest eigenvalues of -0.551, -0.531, -0.398, -0.383, and − 0.298 respectively in PC1. A strong positive correlation (r = 0.81***) was observed between leaf width and leaf length. Genome-wide SNP markers were generated using the DArTseq Genotype by Sequencing approach. Polygenic analysis using 19,467 SNPs identified four distinct genetic groups within cassava population. Phylogenetic and PCA analyses yielded consistent results. Molecularly, PCA revealed that the first three PCs explained 15%, 4.74%, and 3.7% of the genetic variation in the cassava population. SNP markers are effective for evaluating genetic diversity and identifying duplicates in the cassava collection. These results have important implications for cassava genebank management, quality control, conservation strategy, germplasm exchange, and future breeding efforts.

Genetic relatedness

in vitro

Manihot esculenta Crantz

morphological descriptors

SNP markers

Cassava (Manihot esculenta Crantz; Family: Euphorbiaceae) is a vital crop in tropical regions, with 302.7 million tons in 2020 globally. In the same year, West Africa produced 100.6 million tonnes of cassava, accounting for more than 33.2% of global production (FAOSTAT, 2023). However, climate change, including rising temperatures, erratic rainfall, and the spread of pests and diseases, threaten crop yields and food security. IITA, (2014) reported Over 1 billion US$ for managing cassava mosaic disease (CMD) and cassava brown streak diseases (CBSD) every year. To maximize the genetic conservation of cassava, in vitro techniques such as slow growth conditions and cryopreservation were employed to support field preservation since they offer a space, cost, and health-saving method for maintaining and exchanging germplasm (Panis et al., 2020; Salgotra & Chauhan, 2023).

During in vitro conservation, cassava plantlets are initially grown in a culture medium for 3–4 weeks at 28–30℃ via meristem, nodal cutting, or embryo culture with 12-hour light cycles. Subsequently, for medium-term storage, plantlets are transferred to cooler conditions (18–24℃) with reduced light intensity, slowing their growth and metabolic activity (Ng & Ng, 1997). Long-term preservation, however, requires cryopreservation, where plantlets are stored at an ultra-low temperature of -196℃, essentially halting all biological activity and ensuring everlasting conservation (Gueye et al., 2012). For Medium term storage, cassava plantlets are multiplied every 6–18 months (Gueye et al., 2012). During those conservation processes, somaclonal variation may occur ( Antwi,2018).

Antwi (2018) reported that during tissue culture, external issues like problems with air conditioning and lighting systems that impact the humidity, temperature, and photoperiod of the culture environment could also cause cultures to fail or affect gene expression and DNA methylation, leading to somaclonal variation. Stress, growth regulators, culture conditions, natural mutations, and human error can all cause genetic variation and duplication in cassava accessions conserved using in vitro techniques (Amugune & Nyaboga, 2017). Stress such as temperature shock, media salinity, and nutrient deficiency can stimulate somaclonal variation. Different levels of growth regulators during media preparation can also lead to somaclonal variation (Kaviani, 2011). Moreover, frequent cassava germplasm exchange among farmers causes name confusion, leading to the same name for different accessions and vice versa (Yéo et al., 2021). Limited descriptors limit the differentiation of "consanguineous" varieties (Duminil & Di Michele, 2009). Morphological descriptors are influenced by environmental factors and developmental phases of plants (Elsafy et al., 2015).

The assessment of genetic diversity and duplicate identification in cassava using morphological descriptors at the tissue culture level and SNP markers expedite resource utilization by enabling researchers and genebanks to fully comprehend cassava genetic diversity, and eliminate duplicates to achieve successful breeding programs. Morphological data are often preferred for preliminary assessments of genetic diversity because they are quick and easy to collect. However, Morphological data are readily influenced by the environment (Asma et al., 2023). Therefore, it calls for a more precise and reliable method of genetic diversity analysis. Molecular markers can evade the limitations of conventional characterization.

In this vein, some molecular markers have been proven useful for assessing genetic diversity in some cassava accessions including techniques such as restriction fragment length polymorphism (RFLPs) (Beeching et al., 1993), amplified fragment length polymorphism (AFLP) (Asante & Offei, 2003), random amplified polymorphic DNA (RAPDs) (Asante & Offei, 2003), simple sequence repeats (SSRs), and simple nucleotide polymorphism (SNPs) (John et al., 2014). The SSR and SNP markers stand out as competitive candidates for diversification studies among these marker systems. Microsatellites, on the other hand, may have limitations due to the presence of stutter bands, which cause ambiguous scoring in patterns without prominent bands, complicating the scoring process (Park et al., 2009), as well as their poor transferability across species (Grattapaglia & Kirst, 2008).

In contrast to microsatellites, single nucleotide polymorphisms (SNPs) are more easily explored per genetic locus. The SNP markers are proven to have lower genotyping error rates, potentials for high throughput analysis, high genomic abundance, locus specificity, co-dominance, and are less costly (Adu et al., 2021). The SNPs are also useful for identifying and detecting specific genetic differences, even within species with limited diversity (Ferri et al., 2010). Beyond the use of conventional procedures alone, the inclusion of SNPs has sped up the genetic diversity studies and improvements in selection methods. This study used morphological descriptors at the tissue culture level and SNP markers to assess the genetic diversity and identify duplicates of in vitro cassava accessions. This information will provide insights into the selection process of the breeding program and the efficient management of cassava genetic resources in genebanks.

A total of 101 in vitro cassava accessions with their passport data (Table 2.1) were collected from the Genetic Resources Centre (GRC) of the International Institute of Tropical Agriculture (IITA), Ibadan-Nigeria in February 2022, and used for this study.

Table 2.1: Number of accessions, their country of origin, and biological status

Country of origin	Breeding materials	Landraces	Unknown	Total number of accessions (TNA)
Nigeria	62	11	10	83
Cameroon	1	5	0	6
Brazil	0	1	0	1
Togo	0	1	0	1
Guinea	1	1	0	2
Ghana	0	4	1	5
Sierra Leone	0	1	0	1
Cape Verde	0	1	0	1
Gambia	0	1	0	1
Total	64	23	11	101

Source: Genetic Resources Center (GRC)-IITA

Morphological characterization

The morphological study was carried out in the tissue culture laboratory of the Genetic Resources Centre (GRC), International Institute of Tropical Agriculture (IITA), Ibadan-Nigeria. The study used a completely Randomized Design (CRD) with four replicates (Plate 2.1-2.5). During media preparation, a solution containing 4.43 g/l Murashige & Skoog, (1962), 100 mg/l Myo-inositol, and 30 g/l of sucrose was mixed using a magnetic stirrer. The pH of the culture medium was adjusted to 5.7 + 0.1 using 0.5M NaOH and 0.5M HCL, and 9.25 g/l of Agar was added to solidify the media. The medium was later heated in the microwave for 30 min to fully dissolve the agar. Five milliliters of the medium was dispensed per well-cleaned glass test tube, covered thoroughly, and placed in racks that had been pre-prepared. To avoid contamination, the medium was sterilized and autoclaved at 121℃ and a pressure of 15 psi for 15 min. Afterwards, the medium was removed from the autoclave and allowed to cool at ambient temperature before being ready for use.

Destructive sampling technique was used to collect in vitro morphological data (plant height, number of roots, root length, leaf width, leaf length, number of leaves, and number of leaf lobes) from 7 days after sub-culturing with a 7-day interval for data collection during the 4 weeks experimental period. Vernier caliper was used to measure the length/height and width of the plants, leaves, and roots. All data were collected from 4 healthy plants chosen randomly per accession. Quantitative morphological characters were subjected to Analysis of Variance (ANOVA). The principal component analysis and biplot as well as the hierarchical cluster were generated based on the Unweighted Pair Group with an arithmetic mean (UPGMA) approach. Using ascending hierarchical clustering (AHC) based on data and Ward's method to display a dendrogram, the structure of morphological changeability was visualized (Karim et al., 2020). Pearson’s correlation coefficient was performed using R Software version 4.3.0 (R Development Core Team, 2023). Heritability (h²_b) for each character was also estimated (Hill & Mackay, 2004)

Molecular characterization using SNP markers

DNA extraction and DArTseq SNP genotyping

All 101 in vitro cassava accessions were cultured in the tissue culture lab. For 4 weeks prior to DNA extraction. A 100 mg of plant tissue per sample was collected and stored at -80℃. The extraction of genomic (gDNA) was performed using cetyltrimethylammonium bromide (CTAB) protocol as described by Doyle (1987). The concentration and purity of DNA (ratio 230/210) were evaluated using a NanoDrop spectrophotometer. Additionally, agarose gel electrophoresis (1% agarose gel concentration) was performed to obtain quality DNA.

Library preparation and Illumina genome sequencing

Fifty microliter quantity of genomic DNA (concentration of 50ng/microliter) of each cassava accessions was stored in a PCR plate and dispatched to Diversity Array Technology in Canberra, Australia, for genotyping by sequencing. The DArTseq^TM technology was employed, utilizing a combination of MseI and PstI restriction enzymes to prepare the genomic representation and subsequent next-generation sequencing as described by Kilian et al. (2012). The SNP calling was performed with DArTdb, (a laboratory management system), and DarTsoft14 (DS14 software of Diversity Arrays Technology P/L)., An automated genotypic data analysis program, was used to analyze the sequenced data. The DArTSeq SNP markers were scored as “0”, “1” for homozygote allele, and “2” for heterozygote.

SNP calling and QC analysis

A total of 27,170 Single-Nucleotide Polymorphism (SNP) markers were discovered in the cassava population. All these SNPs were aligned on the cassava reference genome version 8.1 (https://phytozome-next.jgi.doe.gov/info/Mesculenta_v8_1). These SNPs were further filtered based on the call rate (CR ≥ 70%), repeatability (RPT ≥ 95%), minor allele frequency (MAF ≥ 0.01), and missing data rate (MDR ≤ 20%) to remove poor quality SNPs and retained 19,467 good SNPs for further use in the genetic studies.

Genetic diversity and population structure analysis

The R Software version 4.3.0 (R Development Core Team, 2023) and TASSEL software (version 5.2.9) were used to explore the genetic diversity and population structure analysis using Principal Component Analysis (PCA) of the in vitro cassava collection. Genetic diversity was explored by calculating the identity-by-state (IBS) genetic distance using the neighbor-joining (NJ) method to develop a phylogenetic tree. In addition to the genetic diversity analysis, Principal Component Analysis (PCA) of the DArTseq markers was performed in Tassel, and its figures were developed using the ggplot package.

Duplicate identification:

Duplicate individual identification among cassava collections was examined. The allele difference between each pair of cassava individuals, was conducted using TASSEL and R-program. Nine technical DNA replicates (genotype same plant DNA twice) were used to make a threshold to confirm duplicates among the cassava accessions. The identity-by-distance (IBS) matrix was calculated, including nine technical duplicates (used as internal control) in TASSEL software. Another identity-by-distant (IBD) method was also used to identify duplicates and cross-check the results from the IBS method in cassava collection using R-package ASRgenomics. A total of 19,467 good SNPs were further filtered using the default setting of function qc.filtering of ASRgenomics package for additional parameters of heterozygosity >0.95 and Fis (inbreeding coefficient) 1 and retained 16,637 SNPs. The kinship diagnostics function was used with filtered 17,637 SNPs in ASRgenomics to consider an individual as a duplicate. The duplicates were identified by setting a threshold based on nine technical duplicates (average mean of IBS/IBD).

Morphological characterization

Variability across the accessions

Analysis of variance revealed significant differences (P < 0.01) of all traits (plant height, number of roots, root length, leaf width, leaf length, number of leaves, and number of leaf lobes) across accessions and weeks. Root length and number of leaf lobes proved significant differences (P < 0.01) across interaction between accessions and weeks, with mean squares of 6.19*** and 0.871***, respectively as shown in Table 3.1.

Table 3.1 Mean squares of morphological traits among 101 in vitro cassava accessions

SOV	DF	LL (cm)	LW (cm)	NL	NR	PH (cm)	RL (cm)	NLL
Accessions	100	0.781***	0.392***	9.110***	13.401***	5.677***	24.721***	1.972***
Weeks	2	33.637***	8.932***	566.141***	575.059***	543.116***	4343.86***	94.003***
Accessions X Weeks	200	0.132	0.045	1.603	2.723	1.123	6.194***	0.871***
Residuals	909	0.126	0.053	1.459	2.733	1.232	3.815	0.484

SOV: Source of variation, DF: degree of freedom, LL: Leaf length, LW: Leaf width, NL: Number of leaves, NR: Number of roots, PH: Plant height, RL: Root length, and NLL: Number of leaf lobes, *** significant: 0.1 % level of probability

Cluster analysis

Cluster analysis was performed using the ward’s minimum clustering method to generate a dendrogram with identified four distinct clusters (I, II, III, and IV) at a 9.5 level of dissimilarity as shown in Figure 3.1. Cluster I contain 17 accessions with two sub-clusters at a 5.0 level of dissimilarity. Cluster II contains 49 accessions with five sub-clusters at a 5.0 level of dissimilarity. Cluster III contains 18 accessions with three sub-clusters at a 5.0 level of dissimilarity. Cluster IV contains 15 accessions with three sub-clusters at a 5.0 level of dissimilarity. The accessions TMe-3373; and TMe-4132 had the most similar phenotypes (≤ 1 of Euclidean distance matrix), while Cluster II had the largest number of accessions (49).

Figure 3. 1 Dendrogram constructed from ward’s minimum cluster analysis to identify genetic relatedness among 101 in vitro cassava accessions using the standardized Euclidean distance matrix.

The first 3 principal components explained 67.26% of the total variation observed among the accessions studied, with eigenvalues ranging from 2.150 to 1.125 as shown in Table 3.2. The first principal component axis (PC1) had an eigenvalue of 2.150, contributing 30.72% of the total variability. All traits negatively contributed to the total variation in PC1. PC2 and PC3 contributed 20.47% and 16.06% of the total variation with eigenvalues of 1.433 and 1.125, respectively. The identical cutoff points at which each selection trait significantly influenced the PC axes were determined as eigenvalues ≥ 0.2. The traits namely leaf length (LL= -0.551), leaf width (LW= -0.531), number of roots (NR= -0.398), plant height (PH= -0.383), and root length (RL = -0.298) loaded PC1. leaf width (LW= 0.431), leaf length (LL= 0.409), root length (RL= -0.569), and number of roots (NR= -0.515), loaded PC2, and number of leaf lobes (NLL= 0.686), plant height (PH = -0.201), and number of leaves (NL= 0.681) loaded PC3. The remaining characters provided few or no discriminatory powers.

Table 3.2 lists the mean, standard deviation, maximum and minimum values, Least Significant Differences (LSD), and heritability (h²_b) of attributes in the study. A high gap was found for a few variables, including root length (RL) and number of roots (NR), according to the results (Table 4.1). Significant differences were noted in traits related to the number of leaf lobes (NLL) and root length (RL), with mean differences (LSD) of 0.569 and 1.595 respectively. The coefficients of variation ranged from 16.062% for NLL to 34.996% for NR. The heritability (h²_b) of 0.83 was for leaf length (LL), 0.867 for leaf width (LW), 0.824 for the number of leaves (NL), 0.558 for the number of leaf lobes (NLL), 0.796 for the Number of roots (NR), 0.786 for plant height (PH), and 0.749 for root Length (RL). The relationship among traits is illustrated in figure 3.2 The association between leaf width and leaf length was highly significant and positive (r=0.81***).

From the principal component biplot (Figure 3.3) of this study, plant height, number of leaves, root length, and number of roots were positively correlated but negatively contributed to the variability observed in the PC1 axis. Leaf width and leaf length were positively correlated but negatively influenced the PC1 axis. The number of leaf lobes and the root length were negatively correlated. The number of leaf lobes strongly influenced PC2, whereas the plant height strongly influenced PC1.

Table 3. 2 Eigenvalues, percentage of cumulative variance, the first three principal component analyses, and descriptive statistics of seven morphological traits.

TRAITS	PC1	PC2	PC3	Mean	Min	Max	SD	CV	P-value	LSD	h²_b
LL	-0.551	0.409	-0.023	1.013	0.567	1.717	0.255	25.179	0.331	0.291	0.830
LW	-0.531	0.431	0.009	0.558	0.225	1.200	0.181	32.363	0.912	0.189	0.867
NL	-0.142	-0.178	0.681	2.965	1.417	5.583	0.871	29.384	0.188	0.986	0.824
NLL	-0.002	0.080	0.686	2.524	1.333	3.250	0.405	16.062	0.000	0.569	0.558
NR	-0.398	-0.515	0.072	3.020	0.917	5.917	1.057	34.996	0.503	1.350	0.796
PH	-0.383	-0.142	-0.201	3.974	2.683	5.508	0.688	17.308	0.790	0.907	0.786
RL	-0.298	-0.569	-0.141	4.948	1.908	9.159	1.435	29.008	0.00	1.595	0.749
Eigenvalue	2.150	1.433	1.125
% variance	30.72	20.478	16.066
Cumulative % variance	30.72	51.198	67.264

CV: coefficient of variations, SD: standard deviation, LSD: Least Significant Difference, h²_b: Heritability, LL: Leaf length, LW: Leaf width, NL: Number of leaves, NR; Number of roots, PH: Plant height, RL: Root length and NLL: Number of leaf lobes.

Figure 3.2 Pearson’s correlation coefficient of the morphological traits of 101 in vitro cassava accessions

Figure 3.3 Principal component biplot showing the variability among accessions and contribution of traits

Molecular characterization

SNP calling, filtering, and heterozygosity

A total of 27,170 SNP markers were discovered genome-wide in the in vitro cassava population, and all generated SNPs were aligned with cassava reference genome v7. After filtering, 19,467 high-quality SNP markers were retained, representing a 28.35% decrease in the original dataset. This filtering step was necessary to ensure that the remaining SNPs were informative and suitable for use in the genetic studies. All polymorphic information content (PIC) ranged from 0.19 to 0.49. with an average of 0.23. the average minor allele frequency (MAF) and average observed heterozygosity were 0.25 and 0.28 respectively. The call rate ranged from 0.7053 to 1.00. with an average of 0.92987. Marker repeatability ranged from 0.95 to 1.0 after filtering.

Figure 3.4: Minor allele frequency (MAF) distribution

Figure 3.5: proportion of missing data distribution

Figure 3.6: Proportion of heterozygous distribution

Genetic Relationship

The genetic diversity among the studied in vitro cassava accessions, as measured by identity by state (IBS) matrix distance, ranged from 0.121 to 0.382. This moderate level of variation suggests the presence of a diverse allelic pool among the germplasm collection. Accessions TMe-1437_T_NGA_Unk and TMe-1230_T_NGA_BR exhibited the lowest genetic distance (0.121), indicating a high degree of similarity at the molecular level due to their close relationships. Conversely, the greatest genetic distance (0.382) observed between TMe-4424-NGR-BR and TMe-2906-CPV-LR highlights substantial allelic divergence in the collection.

The neighbor-joining (NJ) phylogenic tree revealed four distinct clusters (I, II, III, and IV) (Figure 3.7.a.). The maximum distance to the root was 0.189, branch length ranged from 0.00035 to 0.17 in the cladogram tree. Cluster I contained 32 accessions, of which 24 were breeding/ research materials, four were landraces, and four were unknown (not known whether are breeding material or landraces) as shown in figure 3.7.b.

Of the 31 accessions in cluster III, 18 were breeding materials and originated from Nigeria, 11 were landraces (five from Nigeria, three from Ghana, and each from Brazil and Gambia), and two were unknown from Nigeria. Of the 18 accessions in cluster II, 13 were breeding materials and originated from Nigeria, four were landraces and originated from Cameroun, and one was unknown and originated from Nigeria. Of the eight accessions in Cluster IV, five were breeding materials originating from Nigeria, two landraces (one each from Guinea and Togo), and one was unknown from Nigeria.

Figure 3.7.a. Unrooted neighbor-joining tree showing the genetic relationships among in vitro cassava accessions based SNP Markers and explained according to their breeding status and country of origin, (NG: Nigeria, GH: Ghana, CMR: Cameroun, SR: Sierra Leone, CPV: cape verde, BR: Brazil, GMB: Gambia, TG: Togo

Figure 3.7.b: Clustering of cassava accessions based on SNPs markers with respect to their countries of origin and breeding status. (BR: Breeding material, LND: Landraces, UNK: Unknown)

Population structure analysis

Principal components analysis (PCA) was conducted to understand the structure of the cassava population. PCA analysis and similar phylogenetic analysis could not make any distinguished group based on their biological status. Proportional genetic variance analysis showed that the first three principal components, PC1, PC2, and PC3, explain 15%, 4.74%, and 3.7% of the genetic variance, respectively (Figure 3.8). The first ten PCs contributed 44.3% of the cumulative genetic variance to the total genetic variance of the in vitro cassava population.

Figure 3.8 Principal component (PC) analysis of the population of 89 in vitro cassava accessions based on 19k SNP. (a) PC analysis for PC1 vs PC3, and (b) Genetic variance explained by first 10 PCs.

Figure 3.9: Genetic variance explained by the cumulative contribution of the first 10 PCs in Principal Component (PC) analysis of the population of 87 in vitro cassava accessions using SNP markers

Duplicate Identification using Identity-by-State and Kinship G-Matrix Distance

As described in Figure 3.10, technical DNA replicates by genotyping the same plant DNA twice. This was performed to assess the accuracy and reproducibility of the genotyping process. The study used the IBS distance as a measure of genetic similarity. It calculates the proportion of genetic markers where two samples share the same alleles (genetic variations).

The threshold for duplicate confirmation is set by calculating the average IBS distance of the nine technical replicates for a single accession, and then adding the standard error (a measure of variability) to this average. This ensures that even slight variations within the replicates are accounted for, reducing the risk of excluding true duplicates.

This study established an average threshold based on these replicates' average genetic similarity (IBS distance). A diagonal low IBS distance (closer to 0) indicates high genetic similarity and an IBS distance of less than 0.125 was used as a cut-off point for identifying duplicates and considered identical, while those above were considered unique accessions.

Table 3.3.a. IBS and IBD distance between technical control cassava accessions

Ind-1	TechRep	IBS-distance	IBD_distance and Similarity
TMe_F-1	TMe_F-1_2nd	0.120	1.58 (0.98)
TMe_F-7	TMe_F-7_2nd	0.117	1.52 (0.98)
TMe_F-80	TMe_F-80_2nd	0.104	1.50 (0.98)
TMe_IV-84	TMe_IV-84_2nd	0.109	1.43 (0.94)
TMe_IV-86	TMe_IV-86_2nd	0.139	1.62 (0.95)
TMe_IV-928	TMe_IV-928_2nd	0.108	1.46 (0.98)
TMe_IV-929	TMe_IV-929_2nd	0.154	1.18 (0.87)
TMe_IV-3338	TMe_IV-3338_2nd	0.142	1.73 (0.98)
TMe_IV-3442	TMe_IV-3442_2nd	0.128	1.58 (0.98)
Average		0.125	1.51 (0.961)

Figure 3.10.a: Identity by state distance between controlled duplicates

Out of 89 samples, one cassava accession (TMe-1437) was identified as duplicate with TMe-1230 ( 0.122 IBS matrix of 89 lines ). This process ensures that the remaining cassava accessions represent unique genetic individuals, improving the study's overall data quality and accuracy. Similarly, two pairs of individuals of cassava accessions (TMe.3398_T & TMe.70_T and TMe.3235_T & TMe.3252_T) were identified as duplicates using an average correlation value >0.961 (Table 3.3a) of IBD-distance. There were 11 pairs of duplicate accessions confirmed, including TMe.1437_T & TMe.1230_T cassava pairs identified in IBS distance if we used an average correlation value >0.95 (Table 3.3b) as a cutoff. This gives better duplicate identification results than a correlation value >0.95.

Table 3.3. b Duplicate list of cassava pairs, including technical control cassava accessions using Kinship G-matrix (IBD-distance).

S.No	Indiv.A	Indiv.B	IBD-Value	Corr*
1	TMe_F.1	TMe_F.1_2nd	1.582674	0.9814242
2	TMe_IV.3338_2nd	TMe_IV.3338	1.733535	0.9811235
3	TMe_IV.3442_2nd	TMe_IV.3442	1.577154	0.980804
4	TMe_F.80	TMe_F.80_2nd	1.499839	0.9795043
5	TMe_F.7	TMe_F.7_2nd	1.524889	0.9780268
6	TMe_IV.928	TMe_IV.928_2nd	1.462965	0.9764166
7	TMe.3398_T	TMe.70_T	1.233879	0.9642543
8	TMe.3235_T	TMe.3252_T	1.574684	0.9611765
9	TMe.3398_T	TMe.4562_T	1.229085	0.9599954
10	TMe.4562_T	TMe.70_T	1.232826	0.9594426
11	TMe.1230_T	TMe.1437_T	1.634677	0.9574284
12	TMe.4593_T	TMe_IV.84_2nd	1.445569	0.9557188
13	TMe.3398_T	TMe.3314_T	1.225516	0.9555749
14	TMe.3398_T	TMe.1572_T	1.223386	0.9546295
15	TMe.3314_T	TMe.70_T	1.226887	0.9531912
16	TMe_IV.86_2nd	TMe_IV.86	1.619672	0.9514253
17	TMe.1572_T	TMe.70_T	1.222898	0.9508044
18	TMe.452_T	TMe_F.1	1.544997	0.9505431

Corr*: Use as 0.95 as a threshold to confirm duplicate pairs of cassava

Figure 3.10.b: Kinship G-matrix reflecting genetic relationship between cassava individuals. The red vertical dash line indicates probable duplicates of cassava individual pairs (based on average IBD-distance of technical control) in the cassava collection.

Morphological Study

The observed variation in morphological traits among the 101 in vitro cassava accessions strongly indicates their diverse genetic nature, suggesting significant potential for enhancing desired traits through plant breeding. Notably, coefficients of variation ranged from 16.062% (The number of leaf lobes) and 34.96% the number of roots. It is in line with the study ofTemegne et al., (2016) who revealed that the number of leaf lobes has the lowest coefficient of variation. Therefore, this trait could serve as a useful marker for identifying and characterizing cassava genotypes.

Highly significant differences in root length (RL) and the number of leaf lobes (NLL) were observed, which are important traits for both leafy vegetable markets and starch production. The observed differences between these cassava accessions are likely due to variations in their genes (Temegne et al., 2016).Vidal et al., (2015) demonstrated that leaves can vary significantly in shape, color, and size of in vitro cassava. A study conducted in Indonesia on 29 cassava accessions found that morphological characteristics such as plant height, leaf lobes characteristics, and petiole orientation are key factors that contribute to high yield in cassava breeding (Diaguna et al., 2022). The variety of leaf characteristics is crucial in leafy vegetable markets, as it influences cassava selection and cultivar identification.

A study examining the genetic and morphological variations induced by tissue culture in tetraploid cotton (Gossypium hirsutum L.) demonstrated substantial variations in the number of leaves and stem length (Sheidai et al., 2008). Hence, the observed variations in RL and NLL among the cassava accessions suggest the potential for developing improved cassava cultivars with enhanced traits for both leafy vegetable markets and starch production.

The results indicated that all traits were highly heritable (0.558–0.867), implying that they did not have much background noise. A parallel study characterized 419 cassava accessions using 12 distinct morphological descriptors, revealing significant variation among accessions and suggesting a high heritability for these traits, as reported by Mezette et al., (2013). Similar results were also reported by Agre et al., (2015). Therefore, the observed variation in morphological traits among in vitro cassava accessions can be used in plant breeding programs.

Significant variation among in vitro cassava accessions due to genetic variation, environmental factors like light, nutrients, and relative humidity, somatic variation, and human error factors through mislabeling, cross-contamination, and mis-documentation has been reported by Emmy et al., (2015); and Smith, (2019). The variation among in vitro cassava accessions is important for researchers to select accessions with desirable morphological characteristics, identify abiotic stress tolerant accessions, and understand the genetic diversity of cassava germplasm collections.

The noticeable morphological distinctions observed among the regenerated cassava accessions imply that the acquired molecular and genetic variations contribute to these differences. These findings underscore the potential of tissue culture for prompting the development of novel morphological traits, potentially encompassing new agronomic features. These traits could be valuable for breeding purposes in the future.

Clustering categorizes accessions into distinct groups based on the similarity among morphological traits. In this study, morphological clustering of cassava accessions revealed four distinct groups, providing insights into the distribution of genetic diversity. Interestingly, the clustering patterns did not align with geographical locations. This observation aligns with the findings of Mtunguja et al., (2017) who also reported a lack of correlation between morphological clustering and geographical origin in cassava accessions. A plausible explanation is that cassava germplasm has been subjected to extensive movement and exchange, leading to a complex distribution of genetic diversity that transcends geographical boundaries.

Some accessions (TMe 3373, TMe 4132, TMe 178, and TMe 892) showed high degrees of similarity (≤ 1.00 of height dissimilarity Euclidean distance). The duplicates in in vitro conservation may be due to different factors, including human error, cross-contamination, genetic drift, genetic instability, and somatic variation. Therefore, it is important to implement quality control measures and best practices for in vitro cassava conservation to minimize the occurrence of duplicates.

Human error is the predominant cause of duplicates in in vitro cassava conservation, accounting for up to 90% of all cases arising at various stages such as sample collection, labeling, and storage (Smith, 2019). This error is often exacerbated by factors such as insufficient training and experience because cassava tissue culture involves intricate processes. Fatigue and carelessness among personnel, compounded by poorly designed procedures and equipment failures, which all contribute significantly to this challenge (FAO, 2010). It is imperative to implement a comprehensive solution. Using a barcode system for meticulous sample tracking, establishing a double-checking mechanism for labeling and storage processes, conducting regular audits to identify and rectify duplicate entries, and providing rigorous training to personnel on the critical importance of accuracy and potential consequences of human error are vital measures to mitigate this issue effectively.

The PCA results indicated that the three first principal components were major contributors to the observed variation, reflecting the complexity of morphological traits among accessions. The morphological traits that attributed more to the variation observed could be considered and utilized in an effective selection for future breeding and crop improvement programs of cassava accessions. The significant correlation between leaf width (LW) and leaf length (LL) underscores the potential interdependence of these traits in cassava growth and development.

The comprehensive assessment presented in this study has profound implications for plant breeding, germplasm conservation, and future research. The in vitro cassava accessions exhibited high genetic variability, which makes the root length (RL) and number of leaf lobes (NLL) desirable parameters for germplasm characterization and selection. The PCA and Pearson's correlation coefficients provided insight into potential trait interactions, assisting in the discovery of essential traits that affect the overall degree of phenotypic variation. Our understanding of the genetic diversity and connections between morphological traits has improved as a result of the combination of ANOVA, PCA, Cluster Analysis, and Pearson's correlation coefficients. This study offers promise for the development of cassava breeding programs and has implications for germplasm conservation.

Genetic diversity and population structure

The combined analysis of morphological and molecular characterization is a valuable method for assessing the genetic diversity within plant species. In this study, Neighbor-joining phylogenetic analysis based on SNP markers confirmed four clusters at a 95% bootstrap threshold. However, the clusters did not reflect the country of origin of the accessions, and accessions from different countries were grouped in the same clusters. These results align with those of an earlier study conducted in Ghana, where three clusters were formed from 109 cassava accessions based on 105 SNP markers, and the clusters did not correspond to the geographical origin of the accessions. (Naa et al., 2020). A possible explanation for the grouping of accessions from various agroecological zones within the various clusters is the high frequency of germplasm exchange among farmers or often use the same name for different accessions of cassava. At the tissue culture level, soma clonal variation and human error are critical factors for this experience (Amugune & Nyaboga, 2017).

Generally, the SNP dendrogram showed long external branches and small internal branches, suggesting that within-cluster variability was greater than between-cluster variability. This suggests that the accessions within each geographical location were highly diverse and that the accessions were not misnamed. Similar findings have been reported in which farmers are familiar with different varieties within their regions (Augusto et al., 2021). According to Ferguson et al., (2021), planting material is exchanged among farmers over a large agroecological area, and landraces lose their original identity as a result of being given other names. This was the case in this study because of the high degree of accession variability within a given region.

The dendrogram revealed that accessions TMe-3373 and TMe-4132, along with TMe-178 and TMe-892, exhibited the lowest height dissimilarity (< 1.0), suggesting that they were morphologically identical or perhaps duplicates. However, SNP analysis unveiled their genetic distinctness, demonstrating that morphological similarity does not necessarily imply genetic identity. This finding aligns with previous studies, such as Benesi, (2010), which observed that despite sharing 100% morphological similarity, the cultivars Mutuvi and Depwete clustered separately when analyzed using an amplified fragment length polymorphism (AFLP) study. Similarly, Pierre et al., (2022) reported instances of accessions exhibiting similar morphological traits but distinct molecular profiles. This may be explained by environmental effects. Hence, integrating both morphological and molecular data is crucial for effective cassava genetic diversification

Duplicate analysis

Genetic characterization using morphological and molecular characterizations revealed that accessions TMe-3235 and TMe-3252 are genetically identical, exhibiting high similarity (Euclidean distance ≤ 4.00 for morphological traits and IBS distance of 0.134 for molecular markers). In contrast, accessions TMe-1437 and TMe-1230, TMe-1128, and TMe-2966, along with TMe-70, TMe-148, TMe-3314, TMe-743, TMe-1074, TMe-2215, TMe-4562, and TMe-1572, were found to be morphologically distinct. However, SNP analysis indicated that their genetic identity was only present at an identity by state (IBS) distance less than 0.15 threshold. This suggests that the observed morphological differences may be due to environmental factors or epigenetic modifications rather than underlying genetic variation as supported by Kaviani, (2011). Additionally, the absence of duplicates at an IBS distance threshold of 0.05 further supports the presence of genetic diversity among the cassava accessions. These findings are consistent with those reported by Naa et al., (2020), who also found no duplicates among 105 cassava landraces analyzed using 195 SNP markers.

The two accessions from Nigeria with the lowest dissimilarity IBS distance (0.121) were TMe-1230, a breeding material, and TMe-1437, an unknown. On the other hand, the landrace from Cape Verde, TMe-2906, and TMe-4424, a breeding material from Nigeria, had the highest dissimilarity IBS distance (0.3826). The average genetic distance across all 89 in vitro cassava accessions was 0.316 (IBS distance).

This study demonstrates the utility of morphological traits as a rapid and efficient method for assessing germplasm diversity in cassava, particularly at the tissue culture level. However, the study also highlights the limitations of morphological descriptors, particularly in distinguishing genetically identical accessions. This underscores the importance of integrating SNP analysis as a complementary approach to morphological characterization for accurate and comprehensive assessment of genetic diversity in cassava germplasm management and crop improvement programs.

This study assessed the genetic diversity of 101 in vitro cassava accessions using quantitative morphological descriptors and SNP markers, identifying duplicates and revealing significant traits for breeding. Root length and the number of leaf lobes were key morphological traits, while cluster and principal component analyses highlighted four genetic clusters. Leaf width and length showed a strong correlation, emphasizing their collective importance.

The SNP markers confirmed significant diversity, with the lowest dissimilarity observed between two Nigerian accessions, suggesting a common ancestry or similar selection pressures, and the highest dissimilarity between a Cape Verde landrace and a Nigerian breeding material, indicating distinct genetic backgrounds. The study highlighted inefficiencies in cassava germplasm banks, with 19.1% duplicates, and recommended the use of SNP markers for more reliable diversity assessment and efficient germplasm management. Multivariate techniques effectively grouped accessions into four clusters, guiding hybridization parent selection and further characterization to confirm identities and optimize conservation efforts.

Author Contribution

Moïse Hubert Byiringiro:• Conceptualization: Contributed to the initial research concept.• Methodology: Developed the research methodology.• Data acquisition: Performed the majority of data collection for both morphological and molecular studies.• Formal analysis: Analyzed the collected data.• Writing - Original Draft: Drafted the initial manuscript.• Writing - Review & Editing: Incorporated feedback from co-authors and addressed reviewer comments.Dr. Esther Uchendu:• Supervision (Morphological): Supervised all aspects of the morphological analysis and guided the study.• Writing - Review & Editing: Reviewed and provided critical feedback on the manuscript, particularly focusing on the morphological aspects.Dr. Rajneesh Paliwal:• Supervision (Molecular): Supervised all aspects of the molecular analysis and provided guidance during the study.• Formal analysis: Analyzed the Molecular data.• Writing - Review & Editing: Reviewed and provided critical feedback on the manuscript, particularly focusing on the molecular aspects.Prof. Michael Abberton:• Funding Acquisition: Secured the financial resources necessary to conduct the research.• Project Administration: Played a key role in managing the overall research project.• Conceptualization: Contributed valuable insights and ideas that shaped the direction of the research.Note: All authors have read and approved the final version of the manuscript

The acknowledgement: This work was supported by Pan African University Life and Earth Sciences Institute (including Health and Agriculture), PAULESI , Oyo State 200284, Nigeria and International Institute of Tropical Agriculture (IITA), Oyo Road PMB 5320, Ibadan, Nigeria

The submitted manuscript is original and has not been previously published or submitted to another journal for consideration. The work presented does not violate any existing copyright laws or infringe upon the rights of others. All authors have reviewed and approved the manuscript. The authors declare that there are no conflicts of interest related to this research.

Ethical approval

Not applicable.

Adu, B. G., Akromah, R., Amoah, S., Nyadanu, D., Yeboah, A., Aboagye, L. M., Amoah, R. A., & Owusu, E. G. (2021). High-density DArT-based SilicoDArT and SNP markers for genetic diversity and population structure studies in cassava (Manihot esculenta Crantz). PLOS ONE, 16(7), e0255290. https://doi.org/10.1371/journal.pone.0255290
Adu, B. G., Yeboah, A., Akromah, R., Bobobee, E., Amoah, S., Kena, A. W., & Amoah, R. A. (2020). Whole genome SNPs and phenotypic characterization of cassava (Manihot esculenta Crantz) germplasm in the semi-deciduous forest ecology of Ghana. Ecological Genetics and Genomics, 17, 100068. https://doi.org/10.1016/j.egg.2020.100068
Agre, A. P., Dansi, A., Rabbi, I. Y., Battachargee, R., Dansi, M., Melaku, G., Augusto, B., Sanni, A., Akouegninou, A., & Akpagana, K. (2015). Agromorphological Characterization of Elite Cassava (Manihot esculenta Crantz) Cultivars Collected in Benin. In International Journal of Current Research in Biosciences and Plant Biology (Vol. 2015, Issue 2). www.ijcrbp.com
Amugune, N., & Nyaboga, E. (2017). MOLECULAR CHARACTERIZATION OF CASSAVA (Manihot esculenta Crantz.) PLANTS REGENERATED THROUGH ORGANOGENESIS AND SOMATIC EMBRYOGENESIS GILBERT ODINDO OSENA (BSc) Signature Date.
Antwi, N. A. (2018). 188937-Article Text-479870-1-10-20190816 (2). Agricultural and Food Science Journal of Ghana. Vol. 11. Special Edition. July 2018, Vol. 11.(Special Edition. July 2018), 938–939.
Asante, I. K., & Offei, S. K. (2003). RAPD-based genetic diversity study of fifty cassava (Manihot esculenta Crantz) genotypes. Euphytica, 131(1), 113–119. https://doi.org/10.1023/A:1023056313776
Asma, Hussain, M., Ali, N., Masood, R., Akbar, N., Shafqat, N., & Shad, N. (2023). Agro-morphological characterization of Pakistani maize accessions using qualitative and quantitative traits. Brazilian Journal of Biology, 84. https://doi.org/10.1590/1519-6984.259729
Augusto, Lopez, L. L., Bohorquez, C. A., & Zhang, X. (2021). Identification of Cassava Varieties in Ex-Situ Collections and Global Farmer’s Fields: An Update from 1990 to 2020. In Cassava - Biology, Production, and Use. IntechOpen. https://doi.org/10.5772/intechopen.99110
Beeching, J. R., Marmey, P., Gavalda, M.-C., Noirot, M., Haysom, H. R., Hughes, M. A., & Charrier, A. (1993). An Assessment of Genetic Diversity Within a Collection of Cassava (Manihot esculenta Crantz) Germplasm Using Molecular Markers. In Source: Annals of Botany (Vol. 72, Issue 6).
Benesi, I. M. H. J. S. N.,M M. L. K. (2010). Characterisation of Malawian cassava germplasm for diversity, starch extraction and its native and modified properties.
Cruz, V. M. V., Kilian, A., & Dierig, D. A. (2013). Development of DArT Marker Platforms and Genetic Diversity Assessment of the U.S. Collection of the New Oilseed Crop Lesquerella and Related Species. PLoS ONE, 8(5), e64062. https://doi.org/10.1371/journal.pone.0064062
Diaguna, R., Suwarto, Santosa, E., Hartono, A., Pramuhadi, G., Nuryartono, N., Yusfiandayani, R., & Prartono, T. (2022). Morphological and Physiological Characterization of Cassava Genotypes on Dry Land of Ultisol Soil in Indonesia. International Journal of Agronomy, 2022. https://doi.org/10.1155/2022/3599272
Doyle, J. J.; D. J. L. (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15., 19, 11–15.
Duminil, J., & Di Michele, M. (2009). Plant species delimitation: A comparison of morphological and molecular markers. Plant Biosystems - An International Journal Dealing with All Aspects of Plant Biology, 143(3), 528–542. https://doi.org/10.1080/11263500902722964
Egea, L. A., Mérida-García, R., Kilian, A., Hernandez, P., & Dorado, G. (2017). Assessment of Genetic Diversity and Structure of Large Garlic (Allium sativum) Germplasm Bank, by Diversity Arrays Technology “Genotyping-by-Sequencing” Platform (DArTseq). Frontiers in Genetics, 8. https://doi.org/10.3389/fgene.2017.00098
Elsafy, M., Garkava-Gustavsson, L., & Mujaju, C. (2015). Phenotypic Diversity of Date Palm Cultivars (Phoenix dactylifera L.) from Sudan Estimated by Vegetative and Fruit Characteristics. International Journal of Biodiversity, 2015, 1–7. https://doi.org/10.1155/2015/610391
Emmy, C., Miriam, K., Oliver, K., Edith, A., & Stephen, K. (2015). Genetic Diversity of Cassava Mutants, Hybrids and Landraces Using Simple Sequence Repeat Markers. American Journal of Experimental Agriculture, 5(4), 287–294. https://doi.org/10.9734/AJEA/2015/7556
FAO. (2010). In vitro conservation of cassava genetic resources. FAO Technical Guidelines.
FAOSTAT. (2022). Food and agriculture organazation of the united nations statistics division.. Https://Www.Fao.Org/Faostat/En/#home.
Ferguson, M. E., Tumwegamire, S., Chidzanga, C., Shah, T., Mtunda, K., Kulembeka, H., Kimata, B., Tollano, S., Stephen, M., Mpayo, E., Mohamedi, S., Kasele, S., Palangyo, E., Armachius, J., Hamad Ali, A., Sichalwe, K., Matondo, D., Masisila, F., Matumbo, Z., … Mkamilo, G. (2021). Collection, genotyping and virus elimination of cassava landraces from Tanzania and documentation of farmer knowledge. PLOS ONE, 16(8), e0255326. https://doi.org/10.1371/journal.pone.0255326
Ferri, L., Perrin, E., Campana, S., Tabacchioni, S., Taccetti, G., Cocchi, P., Ravenni, N., Dalmastri, C., Chiarini, L., Bevivino, A., Manno, G., Mentasti, M., & Fani, R. (2010). Application of multiplex single nucleotide primer extension (mSNuPE) to the identification of bacteria: The Burkholderia cepacia complex case. Journal of Microbiological Methods, 80(3), 251–256. https://doi.org/10.1016/j.mimet.2010.01.008
Grattapaglia, D., & Kirst, M. (2008). Eucalyptus applied genomics: from gene sequences to breeding tools. New Phytologist, 179(4), 911–929. https://doi.org/10.1111/j.1469-8137.2008.02503.x
Gueye, B., Adeyemi, A., Debiru, M., Akinyemi, B., Olagunju, M., Okeowo, A., Otukpa, S., & Dumet, D. (2012). Standard Operation Procedures (SOP) for IITA in vitro genebank.
Hill, W. G., & Mackay, T. F. C. (2004). D. S. Falconer and Introduction to Quantitative Genetics. Genetics, 167(4), 1529–1536. https://doi.org/10.1093/genetics/167.4.1529
IITA. (2014). IITA Bulletin, Issue 2215. Nigeria: International Institute of Tropical Agriculture.. http://www.iita.org/c/document_library/get_ file?p_l_id = 3299855&folderId = 4274004&name = DLFE-7712.pdf
John, N. N., Francis, N. W., Peter, N., Henry, O., Hannington, O., David, K. L., & Miriam, G. K. (2014). Genetic diversity study of Kenyan cassava germplasm using simple sequence repeats. African Journal of Biotechnology, 13(8), 926–935. https://doi.org/10.5897/AJB12.2426
Kamanda, I., Blay, E. T., Asante, I. K., Danquah, A., Ifie, B. E., Parkes, E., Kulakow, P., Rabbi, I., Conteh, A., Kamara, J. S., Mensah, H. K., Whyte, J. B. A., & Sesay, S. (2020). Genetic diversity of provitamin-A cassava (Manihot esculenta Crantz) in Sierra Leone. Genetic Resources and Crop Evolution, 67(5), 1193–1208. https://doi.org/10.1007/s10722-020-00905-8
Karim, K. Y., Ifie, B., Dzidzienyo, D., Danquah, E. Y., Blay, E. T., Whyte, J. B. A., Kulakow, P., Rabbi, I., Parkes, E., Omoigui, L., Norman, P. E., & Iluebbey, P. (2020). Genetic characterization of cassava (Manihot esculenta Crantz) genotypes using agro-morphological and single nucleotide polymorphism markers. Physiology and Molecular Biology of Plants, 26(2), 317–330. https://doi.org/10.1007/s12298-019-00740-x
Kaviani, B. (2011). Conservation of plant genetic resources by cryopreservation. In AJCS (Vol. 5, Issue 6).
Kouakou, D. A., Kevin,;, Koffi, K., Chia, ;, Angui, M. V., Any, ;, Komenan, O., Irié, A., & Bi, Z. (2023). Agro morphological variability of cassava varieties cultivated in five regions of Côte d’Ivoire based on quantitative traits. Journal of Applied Biosciences, 181, 18962–18973. https://doi.org/10.35759/JABs.181.7
Mezette, T. F., Blumer, C. G., & Veasey, E. A. (2013). Morphological and molecular diversity among cassava genotypes. Pesquisa Agropecuária Brasileira, 48(5), 510–518. https://doi.org/10.1590/S0100-204X2013000500007
Monteros-Altamirano, Á., Tapia, C., Paredes, N., Alulema, V., Tacán, M., Roura, A., Lima, L., & Sørensen, M. (2021). Morphological and Ecogeographic Study of the Diversity of Cassava (Manihot esculenta Crantz) in Ecuador. Agronomy, 11(9), 1844. https://doi.org/10.3390/agronomy11091844
Moura, E. F., Farias Neto, J. T. de, Sampaio, J. E., Silva, D. T. da, & Ramalho, G. F. (2013). Identification of duplicates of cassava accessions sampled on the North Region of Brazil using microsatellite markers. Acta Amazonica, 43(4), 461–467. https://doi.org/10.1590/S0044-59672013000400008
Mtunguja, M. K., Ranjan, A., Laswai, H. S., Muzanila, Y., Ndunguru, J., & Sinha, N. R. (2017). Genetic diversity of farmer-preferred cassava landraces in Tanzania based on morphological descriptors and single nucleotide polymorphisms. Plant Genetic Resources, 15(2), 138–146. https://doi.org/10.1017/S1479262115000453
Naa, R., Prempeh, A., Akwasi Manu-Aduening, J., Quain, M. D., Kwasi Asante, I., Offei, S. K., & Danquah, E. Y. (2020). African Journal of Biotechnology Assessment of genetic diversity among cassava landraces using single nucleotide polymorphic markers. 19(6), 383–391. https://doi.org/10.5897/AJB2018.16420
Ng, S. Y. C., & Ng, N. Q. (1997). Cassava in vitro germplasm management at the International Institute of Tropical Agriculture (Vol. 2, Issue 2).
Panis, B., Nagel, M., & Van den houwe, I. (2020). Challenges and Prospects for the Conservation of Crop Genetic Resources in Field Genebanks, in In Vitro Collections and/or in Liquid Nitrogen. Plants, 9(12), 1634. https://doi.org/10.3390/plants9121634
Park, Y.-J., Lee, J. K., & Kim, N.-S. (2009). Simple Sequence Repeat Polymorphisms (SSRPs) for Evaluation of Molecular Diversity and Germplasm Classification of Minor Crops. Molecules, 14(11), 4546–4569. https://doi.org/10.3390/molecules14114546
Perez, R. D., & Villamayor, F. G. (1984). Relationships among specific gravity, dry matter and starch contents of cassava roots. National Science and Technology Authority Technology Journal, 9(2), .
Pierre, N., Wamalwa, L. N., Muiru, W. M., Simon, B., Kanju, E., Ferguson, M. E., Ndavi, M. M., & Tumwegamire, S. (2022). Genetic diversity of local and introduced cassava germplasm in Burundi using DArTseq molecular analyses. PLOS ONE, 17(1), e0256002. https://doi.org/10.1371/journal.pone.0256002
Plucknett, D. L., Phillips, T. P. and, & Kagbo, R. B. (2000). A Global Development Strategy for Cassava: Transforming a Traditional Tropical Root Crop Spurring Rural Industrial Development and Raising Incomes for the Rural Poor.
R Development Core Team. (2023). A language and environment for statistical computing, reference index version 4.3.1 (4.3.0). R Foundation for Statistical Computing. http://www.R-project.org
Salgotra, R. K., & Chauhan, B. S. (2023). Genetic Diversity, Conservation, and Utilization of Plant Genetic Resources. Genes, 14(1), 174. https://doi.org/10.3390/genes14010174
Sheidai, M., Yahyazadeh, F., Farahanei, F., & Noormohammadi, Z. (2008). Genetic and morphological variations induced by tissue culture in tetraploid cotton (Gossypium hirsutum l.). In Acta Biol Szeged (Vol. 52, Issue 1). http://www.sci.u-szeged.hu/ABSArticle
Smith, P., D. S., E. J., K. S., & G. L. (2019). Human error in in vitro cassava conservation: A review of causes and mitigation strategies. Frontiers in Plant Science, 10(1286).
Temegne, N. C., Ngome Ajebesone, F., & Fotso Kuate, A. (2016). Influence de la composition chimique du sol sur la teneur en éléments nutritifs et le rendement du manioc (Manihot esculenta Crantz, Euphorbiaceae) dans deux zones agro-écologiques du Cameroun.International Journal of Biological and Chemical Sciences, 9(6), 2776. https://doi.org/10.4314/ijbcs.v9i6.21
van Treuren, R., & van Hintum, T. J. L. (2003). MARKER-ASSISTED REDUCTION OF REDUNDANCY IN GERMPLASM COLLECTIONS: GENETIC AND ECONOMIC ASPECTS. Acta Horticulturae, 623, 139–149. https://doi.org/10.17660/ActaHortic.2003.623.15
Vidal, M., Vieira, L. J., Ferreira, C. F., Souza, F. V.D., Souza, A. S., & Ledo, C. A. S. (2015). Genetic fidelity and variability of micropropagated cassava plants (Manihot esculenta Crantz) evaluated using ISSR markers. Genetics and Molecular Research, 14(3), 7759–7770. https://doi.org/10.4238/2015.July.14.2
Yéo, E. F., J-L Amoakon, W., Pita, J. S., Musembi Mutuku, J., Kouassi, M. K., Yao, N., Otron, D. H., Shah, T., L Vanié-Leabo, L. P., H Kpahé, K. M., Sié, R., Sorho, F., Koné, D., N, S.-P. A., Kouassi, N. K., & Ferguson, M. E. (2021). A panel of single nucleotide polymorphism (SNP) markers identifies potential duplicates 2 in cassava (Manihot esculenta Crantz) varieties from Côte d’Ivoire. https://doi.org/10.1101/2021.05.24.445412
Zinga, I., Chiroleu, F., Valam Zango, A., Ballot, C. S. A., Harimalala, M., Kosh Komba, E., Yandia, P. S., Semballa, S., Reynaud, B., Lefeuvre, P., Lett, J.-M., & Dintinger, J. (2016). Evaluation of Cassava Cultivars for Resistance to Cassava Mosaic Disease and Yield Potential in Central African Republic. Journal of Phytopathology, 164(11–12), 913–923. https://doi.org/10.1111/jph.12511

No competing interests reported.

Assessing Genetic Relatedness of In Vitro Cassava Accessions Using Quantitative Morphological Traits and DarTseq Single Nucleotide Polymorphism (SNP) Markers

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods

3. Results

4. Discussion

5. Conclusion

Declarations

Author Contribution

References

Additional Declarations

Status:

Version 1