Life Sciences & Biotechnology
Title : | Machine learning approach to identify the superior haplotypic combinations for higher grain zinc content with higher yield |
Area of research : | Life Sciences & Biotechnology |
Principal Investigator : | Dr. Pallavi Sinha, International Rice Research Institute, Hyderabad, Telangana |
Timeline Start Year : | 2022 |
Timeline End Year : | 2024 |
Contact info : | p.sinha@irri.org |
Equipments : | High-Performance Work Station |
Details
Executive Summary : | Micronutrient deficiencies afflict billions of people throughout the world which is particularly prevalent in developing countries where rice is widely consumed as a staple food. Compared to adults, infants, children, adolescents, pregnant and lactating women have increased requirements for zinc and thus, are at increased risk of Zn depletion (Roohani et al. 2013). Therefore, the development of new high yielding rice varieties containing an increased grain zinc concentration are required to alleviate the ill-effects of malnutrition, especially in children and pregnant women. However, development of such varieties are challenging due to a negative correlation between high grain zinc content and higher grain yield (Sagare et al. 2020). Also, genotype x environment interactions plays crucial role in grain zinc content. Several genes for high grain Zn contents and grain yield have been reported previously. However, there is a limited access to the original donors of the targeted traits from countries in which these genes have been identified. The non-availability of these donors limit the success of breeding programs in order to breed the varieties for the targeted traits. Recently we have proposed an approach called “superior haplotype-based breeding” to identify an alternate donor(s) which can be utilized into the breeding programs. We have demonstrated the approach in two crops, rice (Abbai et al. 2019) and pigeonpea (Sinha et al. 2020). In these studies we have identified the superior haplotypes of the already cloned and characterise genes as well as for newly identified marker-trait associations. While identification of superior haplotypes and their combination, we realised another crucial factor while selecting an appropriate combination of the haplotypes for the breeding programs is the epistatic interaction between and among the superior haplotypes. These epistatic interaction in and between the genes/haplotypes play important role in the expression of the phenotype and need to understand before utilisation into the breeding programs. Recent advances in machine learning-based automated high-throughput phenotyping strategies offer opportunities for documenting precise and accurate phenotypic data. For instance, if we identify superior haplotype(s) and validate them at multiple locations, with the advances in modelling and simulation studies, the information on superior haplotypes of selected key genes could be utilised to simulate and develop ‘digital rice’. Such predicted information will be of great support to design appropriate breeding programs to achieve the goal of tailored rice. Keeping these in view, the proposed work could fasten the transition from breeding to designing crops that are tailored to suit the current and future food-nutritional demands. |
Total Budget (INR): | 28,43,160 |
Organizations involved