System and method for assessing preterm delivery risk for pregnant subject is disclosed. Existing preterm delivery risk assessment methods provide results in late second or third trimester of pregnancy, so little time is available for medical advice. Presently disclosed method and system predict preterm delivery risk within 15 weeks of pregnancy. Microbiome characterization data obtained from microbiome sample from pregnant subject. ‘Microbial taxonomic abundance profile’ generated from microbiome characterization data, contains abundance values of microbes present in the microbiome sample. ‘Taxonomic Composition Skew’ value, and distribution characteristic value for ‘microbial taxonomic abundance profile’, quantifying biases in abundance values of microbes from the microbial taxonomic abundance profile, is computed. Risk of preterm delivery is determined based on the distribution characteristic value or ‘taxonomic composition skew’ value of the set DSR, wherein the set DSR comprises values quantifying biases in the abundance values of microbes from the microbiome sample.