Search

Large plant breeding populations are traditionally a source of novel allelic diversity and are at the core of selection efforts for elite material. Finding rare diversity requires a deep understanding of biological interactions between the genetic makeup of one genotype and its environmental conditions. Most modern breeding programs still rely on linear regression models to solve this problem, generalizing the complex genotype by phenotype interactions through manually constructed linear features. However, the identification of positive alleles vs. background can be addressed using deep learning approaches that have the capacity to learn complex nonlinear functions for the inputs. Machine learning (ML) is an artificial intelligence (AI) approach involving a range of algorithms to learn from input data sets and predict outcomes in other related samples. This paper describes a variety of techniques that include supervised and unsupervised ML algorithms to improve our understanding of nonlinear interactions from plant breeding data sets. Feature selection (FS) methods are combined with linear and nonlinear predictors and compared to traditional prediction methods used in plant breeding. Recent advances in ML allowed the construction of complex models that have the capacity to better differentiate between positive alleles and the genetic background. Using real plant breeding program data, we show that ML methods have the ability to outperform current approaches, increase prediction accuracies, decrease the computing time drastically, and improve the detection of important alleles involved in qualitative or quantitative traits.

Spring frost poses a challenge for all major crops and, in the case of Lupinus mutabilis (Andean lupin) can cause severe damage or even total loss of the crop. Within the LIBBIO project consortium, we conducted a series of experiments in order to develop a suitable protocol for screening lupin germplasm under frost-simulation conditions. Four lupin accessions, one Lupinus albus and three Andean lupins were used in the experiments (L. albus Mihai, L. mutabilis LIB 220, LIB 221, LIB 222). Seedlings at four developmental stages were challenged with five different levels of ‘frost’ stress from low (−2 °C) to high (−10 °C). Notably, young seedling (cotyledons just breaking through the soil surface) showed little evidence of frost damage for temperatures down to −6 °C. At −8 °C, however, damage was evident, suggesting a cold tolerance threshold occurs at this temperature. Interestingly, for later developmental stages, when the first and second leaves were visible, notable differences were observed starting at −6 °C. The results indicate that the plant growth stage is an important parameter when screening for frost tolerance in germplasm. Overall, by identifying Andean lupin genotypes adapted to high abiotic stress factors, farmers will be able to use it as a reference crop with potentially a commercial interest from the food sector, or cosmetics, and biofuel industries.

The intensity of selection, inadvertent pathways of domestication, and the influence of climate change collectively amplify the frequency of detrimental alleles. This phenomenon, in turn, triggers genetic drift, leading to an in-advertent decline in the diversity of plant genetic resources. Genetic variability is of utmost importance for a species’ adaptability and overall fitness. Our investigation centres on Triticum germplasm within the agroclimatic conditions of the Suceava Gene Bank, encompassing 2021–2022 field trials. Comprehensive morphophysiological characterisation was conducted across 200 cultivated varieties, spanning three Triticum species (T. aestivum, T. turgidum and T. monococcum). Initially, the data underwent meticulous processing involving the computation of amplitude of variation, variance (s2), standard deviation (√s), and coefficients of variation (s%) for three pivotal agronomical traits: plant height, spikelets per spike, and total seeds per spike. Furthermore, an extensive cluster analysis was performed, encompassing multiple vital plant descriptors. The findings unveiled a remarkable dispersion of data, with standard deviation, amplitude of variation, and coefficient of variation collectively indicating substantial variability among the cultivated varieties. Within the same population, an intriguing observation emerged; of the 200 genotypes analysed, 83 exhibited immunity to Septoria tritici. Delving deeper into the statistical analysis, we identified two primary clusters within the population. Overall, a significant proportion of this germplasm showcased elevated phenotype scores, rendering them well-suited for further exploration as foundational material in pre-breeding initiatives.

Tilia tomentosa varieties were analyzed using Random Amplified Polymorphic DNA (RAPD) primers to determine genetic diversity. In this study, six samples provided by the Botanical Garden “Anastasie Fătu” of Iasi, Romania were compared to a variety collected from an ecosystem in Giessen, Germany. A total of 91 bands were obtained from nine RAPD primers, 79 (86.8%) of which were polymorphic. Marker data was used for a principal component analysis that showed clear differentiation among Romanian and German samples. Furthermore, Romanian genotypes were clustered together in the principal component analysis. PCA analysis was followed by an Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis which revealed two major groups. Samples from the Romanian ecosystem showed high genetic similarity. Generally, samples did not separate from each other in the UPGMA analysis, excepting the German sample.