Rice (Oryza sativa) is one of the most important crop species and a staple food for more than half of the world’s population. However, rice is highly salt-sensitive and its yield is affected by salinity-alkalinity. Till today, tremendous efforts have been made to develop tolerant rice cultivars aiming at sustaining or improving rice productivity. However, underlying molecular mechanism of salt stress responses has not yet been very clear. In this project, the major goal is to establish a statistical model, helping us identify genes or pathways related to the ability of salt tolerance. At the molecular level, co-expression network based on RNA-sequencing data was constructed to cluster genes as biologically meaningful modules. At the phenotypic level, sodium-potassium ratio was used as an indicator of the ability of salt tolerance. To build a bridge between two levels, variable selection methods, were applied to identify biological modules that have a large contribution to the observed phenotype.