Fig. 1: Flow chart of method design and prediction model development.

The study was based on a sampling of 337 early-stage breast cancer survivors (ESBCS) with different levels of taxane-induced peripheral neuropathy (TIPN), divided into a training cohort (70%) and a test cohort (30%). Blood samples from the survivors were whole-exome sequenced. Building and validating logistic regression models was based on literature data, single nucleotide variants (SNVs) association analysis, gene/region association analysis, over-representation analysis, and Variable importance of the results from the whole-exome results and lastly by a combination. The figure was created in BioRender.com.