Table 2.

Likelihood ratio test of a two or three component mixture model of HSF scores of HER2-negative tumor cells

Test of fits	Chi-squared	P value
2 vs 1	66.92	1.8e−14
3 vs 2	54.59	8.7e−12

Statistical analysis using a likelihood ratio test reveals that three-component mixture model significantly improves goodness-of-fit compared to a two-component mixture model, where components refer to density curves of CELx HSF test scores shown in Fig. 6a