Computational protocol: Identification of Quantitative Trait Loci Conditioning the Main Biomass Yield Components and Resistance to Melampsora spp. in Salix viminalis × Salix schwerinii Hybrids

Similar protocols

Protocol publication

[…] Biometric parameters, including the height and diameter of the main stem (plant height (ph) and steam diameter (sd)) and the number of shoots per plant (nos), were measured after each growing season. Quantitative trait loci were identified based on measurements of 3-year-old plants growing on a 4-year-old stump. The empirical distribution of the analyzed traits was checked for normality in the Shapiro-Wilk W-test []. The results were processed by one-way ANOVA. The significance of differences between means was analyzed by Tukey’s HSD test, a multiple comparisons procedure, at p ≤ 0.05. All calculations were performed in Statistica v. 12.5 (Statistica, Tulsa, OK, USA) []. [...] The resistance of the tested plants (P5 population) was evaluated on leaf discs with a diameter of 20 mm. Leaf discs were placed on square Petri plates with 25 compartments (Sterilin Limited, ThermoFisher Scientific, Cambridge, UK) lined with filter paper segments (Whatman 3MM, GE Healthcare, Maidstone, UK). Leaf discs representing different genotypes were inoculated with four M. larici-epitea isolates (Mle1–Mle4), in five technical replications and two biological replications each. The suspension of freshly propagated spores at a concentration of 1–2 × 105/mL water, containing 0.004% of the Tween 20 detergent, was placed on Petri plates (10 cm × 10 cm) in the amount of 1 mL per plate with the use of an air brush with a 0.35 mm nozzle (Air Brush Kit, EW-6000B, 0.2 mm, Jadar Model, MAR, Warsaw, Poland). The viability and germination capacity of fungal spores were checked under the light microscope after 24 and 48 h of incubation in 0.5% aqueous agar solution. The severity of fungal infection was evaluated in the ImageJ program (imagej.net, National Institutes of Health, Bethesda, MD, USA) for image processing and analysis. The number and surface area of uredinia were determined, and the results were used to calculate leaf area colonized by fungi. The evaluation was done 13 days post inoculation (13 dpi). The analysis relied on an infected area on the leaf disc, which was calculated by multiplying the number of uredinia by their surface area. If the normal distribution hypothesis for this trait was rejected, data were transformed by the method proposed by Bliss [] [...] Linkage analysis was performed in the R/QTL [], using the procedure described in “Genetic map construction with R/QTL” by Broman and Sen []. Data were prepared for mapping by excluding missing parent alleles, duplicate markers, markers with call rates of less than 0.75, and markers not exhibiting Mendelian segregation. The missing data for the analyzed individuals did not exceed 10%, and all data were included in the analysis which covered a total of 463 markers and 79 individuals. Markers were assigned to linkage groups for analysis as a phase-known four-way cross [] in R/QTL software (University of Wisconsin-Madison, Madison, WI, USA) using the formLinkageGroups function (max.rf. = 0.35 min, LOD = 5). Markers were ordered, rippled, and re-ordered according to pairwise recombination fractions, LOD scores and linkage group length.The QTLs responsible for biomass yield-related traits and leaf area infected by fungi were identified by maximum-likelihood interval mapping with the use of the EM algorithm [] in R/QTL software using the procedure described by Broman and Sen []. QTL genotype probability was calculated using the calc.genoprob function with a step size of 1 cM. Simple interval mapping analyses of each separate trait were first performed to detect potential QTL positions using the scanone function. Genome-wide LOD significance thresholds were calculated based on 1000 permutations at 0.05 α-value level []. For each trait, two-dimensional genome scans were performed for the two-QTL model (scantwo function) to identify successive QTLs, and their location on the genetic map was optimized (makeQTL, fitQTL, refineQTL and addQTL functions). Interval mapping was performed by calculating the 95% Bayesian credible interval [] with the bayesint function in R/QTL software.The percentage variability in a phenotypic trait explained by a given single QTL was assessed using the fitqtl function. The identified QTLs were mapped in the MapChart 2.3 application (Kyazma BV, Wageningen, The Netherlands) []. […]

Pipeline specifications