Table 1.
Important chemical features for regression models. Top three physicochemical features for the viral targets with known bioassay activities.
| Feature | Target | Description |
|---|---|---|
| GATS5s | ABCC1 | Geary autocorrelation of lag 5 weighted by I-state |
| RDF055m | ABCC1 | Radial Distribution Function - 055/weighted by mass |
| SpMax_B(s) | ABCC1 | leading eigenvalue from Burden matrix weighted by I-State |
| CATS2D_08_AA | BRD2 | CATS2D Acceptor-Acceptor at lag 08 |
| RDF035s | BRD2 | Radial Distribution Function - 035/weighted by I-state |
| SpDiam_X | BRD2 | spectral diameter from chi matrix |
| HATS8p | BRD4 | leverage-weighted autocorrelation of lag 8/weighted by polarizability |
| R5i+ | BRD4 | R maximal autocorrelation of lag 5/weighted by ionization potential |
| RDF035m | BRD4 | Radial Distribution Function - 035/weighted by mass |
| Eig02_EA(bo) | CSNK2A2 | eigenvalue n. 2 from edge adjacency mat. weighted by bond order |
| Eig05_EA(bo) | CSNK2A2 | eigenvalue n. 5 from edge adjacency mat. weighted by bond order |
| SpMax2_Bh(m) | CSNK2A2 | largest eigenvalue n. 2 of Burden matrix weighted by mass |
| CATS2D_04_AA | CSNK2B | CATS2D Acceptor-Acceptor at lag 04 |
| SHED_DN | CSNK2B | SHED Donor-Negative |
| SpMin1_Bh(m) | CSNK2B | smallest eigenvalue n. 1 of Burden matrix weighted by mass |
| DISPm | DCTPP1 | displacement value/weighted by mass |
| HATS7u | DCTPP1 | leverage-weighted autocorrelation of lag 7/unweighted |
| Mor31s | DCTPP1 | signal 31/weighted by I-state |
| MATS1e | DNMT1 | Moran autocorrelation of lag 1 weighted by Sanderson electronegativity |
| Mor23m | DNMT1 | signal 23/weighted by mass |
| TDB06u | DNMT1 | 3D Topological distance based descriptors - lag 6 unweighted |
| GATS4m | GFER | Geary autocorrelation of lag 4 weighted by mass |
| Mor14m | GFER | signal 14/weighted by mass |
| R5i | GFER | R autocorrelation of lag 5/weighted by ionization potential |
| DISPp | HDAC2 | displacement value/weighted by polarizability |
| IC2 | HDAC2 | Information Content index (neighborhood symmetry of 2-order) |
| P_VSA_MR_5 | HDAC2 | P_VSA-like on Molar Refractivity, bin 5 |
| F04[C–C] | IMPDH2 | Frequency of C - C at topological distance 4 |
| HOMA | IMPDH2 | Harmonic Oscillator Model of Aromaticity index |
| VE1_B(s) | IMPDH2 | coefficient sum of the last eigenvector (absolute values) from Burden matrix weighted by I-State |
| Eig02_AEA(dm) | ITGB1 | eigenvalue n. 2 from augmented edge adjacency mat. weighted by dipole moment |
| SHED_AA | ITGB1 | SHED Acceptor-Acceptor |
| SpMax2_Bh(s) | ITGB1 | largest eigenvalue n. 2 of Burden matrix weighted by I-state |
| F10[C–N] | MARK2 | Frequency of C - N at topological distance 10 |
| nPyrroles | MARK2 | number of Pyrroles |
| SaaNH | MARK2 | Sum of aaNH E-states |
| max_conj_path | MARK3 | maximum number of atoms that can be in conjugation with each other |
| SaaNH | MARK3 | Sum of aaNH E-states |
| VE1_H2 | MARK3 | coefficient sum of the last eigenvector (absolute values) from reciprocal squared distance matrix |
| GATS3s | NSD2 | Geary autocorrelation of lag 3 weighted by I-state |
| HOMA | NSD2 | Harmonic Oscillator Model of Aromaticity index |
| Mor16s | NSD2 | signal 16/weighted by I-state |
| H7m | PABPC1 | H autocorrelation of lag 7/weighted by mass |
| JGI7 | PABPC1 | mean topological charge index of order 7 |
| P_VSA_MR_2 | PABPC1 | P_VSA-like on Molar Refractivity, bin 2 |
| GATS4m | PLAT | Geary autocorrelation of lag 4 weighted by mass |
| Mor04s | PLAT | signal 04/weighted by I-state |
| R6p+ | PLAT | R maximal autocorrelation of lag 6/weighted by polarizability |
| nPyrroles | PRKACA | number of Pyrroles |
| RDF040v | PRKACA | Radial Distribution Function - 040/weighted by van der Waals volume |
| SpMin3_Bh(m) | PRKACA | smallest eigenvalue n. 3 of Burden matrix weighted by mass |
| Eig02_EA(bo) | PSEN2 | eigenvalue n. 2 from edge adjacency mat. weighted by bond order |
| nArX | PSEN2 | number of X on aromatic ring |
| VE1sign_D/Dt | PSEN2 | coefficient sum of the last eigenvector from distance/detour matrix |
| SHED_DL | PTGES2 | SHED Donor-Lipophilic |
| VE2sign_G | PTGES2 | average coefficient of the last eigenvector from geometrical matrix |
| VE3sign_G | PTGES2 | logarithmic coefficient sum of the last eigenvector from geometrical matrix |
| CATS3D_08_AL | RIPK1 | CATS3D Acceptor-Lipophilic BIN 08 (8.000–9.000 Å) |
| MATS5i | RIPK1 | Moran autocorrelation of lag 5 weighted by ionization potential |
| VE3sign_RG | RIPK1 | logarithmic coefficient sum of the last eigenvector from reciprocal squared geometrical matrix |
| BLTA96 | SIGMAR1 | Verhaar Algae base-line toxicity from MLOGP (mmol/l) |
| F10[C–C] | SIGMAR1 | Frequency of C - C at topological distance 10 |
| TPSA(Tot) | SIGMAR1 | topological polar surface area using N,O,S,P polar contributions |
| Eig01_AEA(dm) | TBK1 | eigenvalue n. 1 from augmented edge adjacency mat. weighted by dipole moment |
| HATS4i | TBK1 | leverage-weighted autocorrelation of lag 4/weighted by ionization potential |
| SdssC | TBK1 | Sum of dssC E-states |
| AROM | VCP | aromaticity index |
| E1m | VCP | 1st component accessibility directional WHIM index/weighted by mass |
| MATS5m | VCP | Moran autocorrelation of lag 5 weighted by mass |
| H5s | ACE2 | H autocorrelation of lag 5/weighted by I-state |
| Mor10m | ACE2 | signal 10/weighted by mass |
| Mor17m | ACE2 | signal 17/weighted by mass |