Variation in cassava (Manihot esculenta Crantz) based on single nucleotide polymorphisms, simple sequence repeats and phenotypic traits

Abstract

English: Science-driven interventions aimed at increasing cassava’s (Manihot esculenta Crantz), productivity will positively impart on communities that primarily depend on it. Data presented in this thesis contributed towards this goal. Specifically, this thesis examined phenotypic variation in cassava germplasm available within selected national breeding programmes, analysed patterns of allele frequency distribution in the cassava germplasm, characterised single nucleotide polymorphisms (SNPs) in the cassava genome and examined variation for two agronomic [root dry matter content (DMC) and harvest index (HI)] and one quality trait (amylose content) in S1 cassava inbreds. The 29 qualitantive traits provided limited discrimination of cassava germplasm from Uganda, Kenya, Tanzania, Rwanda, Democratic Republic of Congo and Madagascar. However, significant differences in root DMC, HI, leaf retention (LR) and root cortex thickess were observed in this germplasm. Highest average DMC was registered in Uganda (37.7%) and lowest in Tanzania (30.1%), with the elite genotypes having a relatively higher DMC than local genotypes. Similary, highest average HI was observed in Uganda (0.60) and lowest in Kenya (0.32). This germplasm (1401 local and elite cassava genotypes) were furthermore genotyped using simple sequence repeat (SSR) markers. Genotypes were assigned to three distinct populations, with no appreciable gene differentiation (FST = 0.089) observed. Most genetic variation (> 89%) was detected within individuals. A total of 26 SNPs were identified from quality sequences of nine genes, giving an estimated frequency of one SNP for every 121 nucleotides. Nucleotide diversity ranged from 7.8 x 10-4 to 5.63 x 10-3. Average haplotype-based polymorphism information content (PIC = 0.414) was higher than for individual SNP (PIC = 0.228). HI and DMC varied both within and between the cassava S1 families. The highest reduction in HI was observed in S1 progeny from Bamunanika (83.3%), while the highest reduction in DMC (23.8%) was observed in S1 progeny from I92/00067. Amylose content varied significantly between cassava families and S1 progeny, with most of the variation recorded within progeny.

Afrikaans: Wetenskaplik-gedrewe intervensies wat ten doel het om cassava (Manihot esculenta Crantz) se produktiwiteit te verbeter, sal ‘n positiewe impak maak op gemeenskappe wat primêr daarop staatmaak. Data wat in hierdie tesis voorgehou is, het tot hierdie doel bygedra. Hierdie tesis het spesifiek die fenotipiese variasie in cassava kiemplasma beskikbaar binne geselekteerde nasionale teelprogramme ondersoek, patrone van alleel frekwensie verspreiding binne die cassava kiemplasma geanaliseer, enkel nukleotied polimorfismes (ENP) in die cassava genoom gekarakteriseer en die variasie vir twee agronomiese [wortel droë materiaal opbrengs (DMO) en oesindeks (OI)] en een kwaliteitseienskap (amilose inhoud) in S1 cassava ingeteelde lyne ondersoek. Die 29 kwalitatiewe eienskappe het tot beperkte diskriminasie van die cassava kiemplasma van Uganda, Kenia, Tanzanië, Rwanda, Demokratiese Republiek van die Kongo en Madagaskar gelei. Betekenisvolle verskille in wortel DMO, OI, blaarretensie (BR) en wortel korteksdikte is egter in hierdie kiemplasma waargeneem. Die hoogste gemiddelde DMO is in Uganda (37.7%) en die laagste in Tanzanië (30.1%) waargeneem, met die elite genotipes wat ‘n relatiewe hoër DMO as die plaaslike genotipes gehad het. Net so is die hoogste gemiddelde OI in Uganda (0.60) en die laagste in Kenia (0.32) waargeneem. Hierdie kiemplasma (1401 plaaslike en elite cassava genotipes) is verder genotipies getipeer deur van eenvoudige volgorde herhalende (EVH) merkers gebruik te maak. Genotipes is in drie duidelike populasies gegroepeer, sonder enige beduidende geen differensiasie (FST = 0.089). Die grootste gedeelte van die genetiese variasie (> 89%) is tussen individue waargeneem. ‘n Totaal van 26 ENPs is van kwaliteitsvolgordes van nege gene geïdentifiseer, wat ‘n geraamde frekwensie van een ENP vir elke 121 nukleotide gee. Nukleotied diversiteit het van 7.8 x 10-4 tot 5.63 x 10-3 gevarieer. Gemiddelde haplotied-gebaseerde polimorfiese informasie inhoud (PII = 0.414) was hoër as vir individuele ENP (PII = 0.228). OI en DMO het beide binne en tussen die cassava S1 families gevarieer. Die grootste verlaging in OI was in S1 nageslag van Bamunanika (83.3%) en die grootste verlaging in DMO (23.8%) in die S1 nageslag van I92/00067 waargeneem. Amilose inhoud het betekenisvol tussen cassava families en S1 nageslag gevarieer, met die grootste gedeelte van die variasie binne die nageslagte.