Prediction of phenotype for multiple traits from multi-omic data

该项目旨在开发更好的方法，以预测个体的基因组序列特征。这种方法将在农业动物、植物和人类身上进行试验。预测公式来源于一个训练数据集，该数据集包含个体样本的特征和基因组序列信息。然后，预测公式可以应用于预测未知个体的特征。这对选择最佳亲本进行农业育种和预测动物、作物和人的未来表型是有用的。该方法利用大量性状的数据来识别具有特定功能的序列变异，并预测受每个变异影响的性状。

This project aims to develop better methods for predicting traits in an individual based on their genome sequence. This method will be tested in agricultural animals and plants and in humans. The prediction formula is derived from a training dataset that has information on the traits and genome sequence of a sample of individuals. The prediction formula can then be applied to predict the trait in individuals where the trait is unknown. This is useful for selecting the best parents for breeding in agriculture and for predicting the future phenotype of animals, crops and people. The proposed method uses data on very many traits to identify sequence variants that have a function and to predict the traits affected by each variant.