Skip to main content

Table 2 Median MCC scores of single task and multi-task QSAR models

From: Analysis of the benefits of imputation models over traditional QSAR models for toxicity prediction

  

Ames

Tox21

Single task

RF

0.526 (0.523–0.527)

0.402 (0.400–0.405)

XGB

0.547 (0.545–0.550)

0.427 (0.422–0.430)

ST-DNN

0.519 (0.508–0.527)

0.417 (0.410–0.426)

Multi-task

XGB-FN

0.529 (0.527–0.531)

0.435 (0.432–0.438)

MT-DNN

0.540 (0.527–0.549)

0.430 (0.423–0.437)

Macau

0.490 (0.484–0.499)

0.321 (0.319–0.323)

  1. Median MCC scores and interquartile range for each technique and dataset on the test set across the 20 random seeds. Before computing the median, the mean across the different assays for a single run was calculated. The best model for each dataset in bold