This is a dataset of whole genome sequencing data (WGS) for 762 Escherichia coli isolates with associate antimicrobial susceptibility testing (AST) meta-data. The dataset was generated from clinical strains retrieved from Liverpool Clinical Laboratories, UK (Liverpool University Hospitals NHS Trust). Isolates were originally isolated between 2017–2021. The dataset was produced to train machine learning models to predict antimicrobial minimum inhibitory concentration from WGS data. The isolates were chosen based on their AST phenotype, to generate a representative sample of important resistance mechanisms. In other words, this is not a representative random sample of E. coli AST phenotypes — rare phenotypes were over-sampled; common phenotypes were under-sampled.