Table 4. MMPA-Generated Key Molecular Transformations Shown to Significantly Affect GN-Activity Scores of Synthetically Generated Dataa.
entry | main moiety | +/– | ± std | repeats | opposite moiety 1 | opposite moiety 2 | opposite moiety 3 |
---|---|---|---|---|---|---|---|
1 | primary amine | addition | 0.42 ± 0.25 | 61 | ether [25%] | carbonyl [20%] | secondary amine [10%] |
2 | lactone | removal | 0.41 ± 0.08 | 35 | secondary amine [14%] | tertiary amine [9%] | ether [6%] |
3 | ester (carboxylate ester) | removal | 0.39 ± 0.16 | 56 | secondary amine [9%] | (ether, primary amine) [7%] | tertiary amine [7%] |
4 | carbonyl | removal | 0.37 ± 0.17 | 96 | thiophene [17%] | primary amine [13%] | aryl chloride [9%] |
5 | nitrile | removal | 0.33 ± 0.13 | 255 | aryl chloride [22%] | (thiophene, ether) [12%] | aryl fluoride [5%] |
6 | thiophene | addition | 0.32 ± 0.15 | 304 | (nitrile, secondary amine) [9%] | aniline [8%] | carbonyl [5%] |
7 | tertiary carboxamide | removal | 0.31 ± 0.17 | 31 | thiophene [42%] | aryl chloride [23%] | secondary carboxamide [19%] |
8 | aryl chloride | addition | 0.28 ± 0.16 | 279 | nitrile [20%] | tertiary amine [6%] | aryl fluoride [4%] |
9 | secondary amine | addition | 0.24 ± 0.13 | 306 | tertiary amine [16%] | ether [3%] | nitrile [3%] |
10 | tertiary amine | removal | 0.22 ± 0.09 | 292 | secondary amine [17%] | (aryl chloride, aryl fluoride) [6%] | alkanol [7%] |
11 | α,β-unsaturated carbonyl | removal | 0.21 ± 0.14 | 33 | thiophene [18%] | secondary amine [15%] | aryl chloride [12%] |
12 | aryl fluoride | addition | 0.20 ± 0.13 | 154 | nitrile [16%] | tertiary amine [12%] | alkanol [3%] |
Listed are the molecular replacements that correlate with a strong increase of the predicted GN activity, most likely due to improved permeability according to our initial curation, by either addition or removal of each specific main moiety (+/−). The top three exchange counterparts in the matched molecular pairs are shown as opposite moieties. denotes the average change of the GN-activity score (limits from 0 to 1), linked to the respective transformations with associated standard deviation. “Repeats” indicates the number of transformations that contain the “main moiety”.