nonzeroexit commited on
Commit
5d02ae6
·
verified ·
1 Parent(s): b9b7438

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +100 -56
app.py CHANGED
@@ -70,62 +70,106 @@ def get_protbert():
70
  # and is intentionally excluded from the input features.
71
  # ---------------------------------------------------------------------------
72
  selected_features = [
73
- '_PolarizabilityC1', '_PolarizabilityC2', '_PolarizabilityC3',
74
- '_SolventAccessibilityC1', '_SolventAccessibilityC2', '_SolventAccessibilityC3',
75
- '_SecondaryStrC1', '_SecondaryStrC2', '_SecondaryStrC3',
76
- '_ChargeC1', '_ChargeC2', '_ChargeC3',
77
- '_PolarityC1', '_PolarityC2', '_PolarityC3',
78
- '_NormalizedVDWVC1', '_NormalizedVDWVC2', '_NormalizedVDWVC3',
79
- '_HydrophobicityC1', '_HydrophobicityC2', '_HydrophobicityC3',
80
- '_PolarizabilityT12', '_PolarizabilityT13', '_PolarizabilityT23',
81
- '_SolventAccessibilityT12', '_SolventAccessibilityT13', '_SolventAccessibilityT23',
82
- '_SecondaryStrT12', '_SecondaryStrT13', '_SecondaryStrT23',
83
- '_ChargeT12', '_ChargeT13', '_ChargeT23',
84
- '_PolarityT12', '_PolarityT13', '_PolarityT23',
85
- '_NormalizedVDWVT12', '_NormalizedVDWVT13', '_NormalizedVDWVT23',
86
- '_HydrophobicityT12', '_HydrophobicityT13', '_HydrophobicityT23',
87
- 'A', 'R', 'N', 'D', 'C', 'E', 'Q', 'G', 'H', 'I',
88
- 'L', 'K', 'M', 'F', 'P', 'S', 'T', 'W', 'Y', 'V',
89
- 'AA', 'AR', 'AN', 'AD', 'AC', 'AE', 'AQ', 'AG', 'AH', 'AI',
90
- 'AL', 'AK', 'AM', 'AF', 'AP', 'AS', 'AT', 'AW', 'AY', 'AV',
91
- 'RA', 'RR', 'RN', 'RD', 'RC', 'RE', 'RQ', 'RG', 'RH', 'RI',
92
- 'RL', 'RK', 'RM', 'RF', 'RP', 'RS', 'RT', 'RW', 'RY', 'RV',
93
- 'NA', 'NR', 'NN', 'ND', 'NC', 'NE', 'NQ', 'NG', 'NH', 'NI',
94
- 'NL', 'NK', 'NM', 'NF', 'NP', 'NS', 'NT', 'NW', 'NY', 'NV',
95
- 'DA', 'DR', 'DN', 'DD', 'DC', 'DE', 'DQ', 'DG', 'DH', 'DI',
96
- 'DL', 'DK', 'DM', 'DF', 'DP', 'DS', 'DT', 'DW', 'DY', 'DV',
97
- 'CA', 'CR', 'CN', 'CD', 'CC', 'CE', 'CQ', 'CG', 'CH', 'CI',
98
- 'CL', 'CK', 'CM', 'CF', 'CP', 'CS', 'CT', 'CW', 'CY', 'CV',
99
- 'EA', 'ER', 'EN', 'ED', 'EC', 'EE', 'EQ', 'EG', 'EH', 'EI',
100
- 'EL', 'EK', 'EM', 'EF', 'EP', 'ES', 'ET', 'EW', 'EY', 'EV',
101
- 'QA', 'QR', 'QN', 'QD', 'QC', 'QE', 'QQ', 'QG', 'QH', 'QI',
102
- 'QL', 'QK', 'QM', 'QF', 'QP', 'QS', 'QT', 'QW', 'QY', 'QV',
103
- 'GA', 'GR', 'GN', 'GD', 'GC', 'GE', 'GQ', 'GG', 'GH', 'GI',
104
- 'GL', 'GK', 'GM', 'GF', 'GP', 'GS', 'GT', 'GW', 'GY', 'GV',
105
- 'HA', 'HR', 'HN', 'HD', 'HC', 'HE', 'HQ', 'HG', 'HH', 'HI',
106
- 'HL', 'HK', 'HM', 'HF', 'HP', 'HS', 'HT', 'HW', 'HY', 'HV',
107
- 'IA', 'IR', 'IN', 'ID', 'IC', 'IE', 'IQ', 'IG', 'IH', 'II',
108
- 'IL', 'IK', 'IM', 'IF', 'IP', 'IS', 'IT', 'IW', 'IY', 'IV',
109
- 'LA', 'LR', 'LN', 'LD', 'LC', 'LE', 'LQ', 'LG', 'LH', 'LI',
110
- 'LL', 'LK', 'LM', 'LF', 'LP', 'LS', 'LT', 'LW', 'LY', 'LV',
111
- 'KA', 'KR', 'KN', 'KD', 'KC', 'KE', 'KQ', 'KG', 'KH', 'KI',
112
- 'KL', 'KK', 'KM', 'KF', 'KP', 'KS', 'KT', 'KW', 'KY', 'KV',
113
- 'MA', 'MR', 'MN', 'MD', 'MC', 'ME', 'MQ', 'MG', 'MH', 'MI',
114
- 'ML', 'MK', 'MM', 'MF', 'MP', 'MS', 'MT', 'MW', 'MY', 'MV',
115
- 'FA', 'FR', 'FN', 'FD', 'FC', 'FE', 'FQ', 'FG', 'FH', 'FI',
116
- 'FL', 'FK', 'FM', 'FF', 'FP', 'FS', 'FT', 'FW', 'FY', 'FV',
117
- 'PA', 'PR', 'PN', 'PD', 'PC', 'PE', 'PQ', 'PG', 'PH', 'PI',
118
- 'PL', 'PK', 'PM', 'PF', 'PP', 'PS', 'PT', 'PW', 'PY', 'PV',
119
- 'SA', 'SR', 'SN', 'SD', 'SC', 'SE', 'SQ', 'SG', 'SH', 'SI',
120
- 'SL', 'SK', 'SM', 'SF', 'SP', 'SS', 'ST', 'SW', 'SY', 'SV',
121
- 'TA', 'TR', 'TN', 'TD', 'TC', 'TE', 'TQ', 'TG', 'TH', 'TI',
122
- 'TL', 'TK', 'TM', 'TF', 'TP', 'TS', 'TT', 'TW', 'TY', 'TV',
123
- 'WA', 'WR', 'WN', 'WD', 'WC', 'WE', 'WQ', 'WG', 'WH', 'WI',
124
- 'WL', 'WK', 'WM', 'WF', 'WP', 'WS', 'WT', 'WW', 'WY', 'WV',
125
- 'YA', 'YR', 'YN', 'YD', 'YC', 'YE', 'YQ', 'YG', 'YH', 'YI',
126
- 'YL', 'YK', 'YM', 'YF', 'YP', 'YS', 'YT', 'YW', 'YY', 'YV',
127
- 'VA', 'VR', 'VN', 'VD', 'VC', 'VE', 'VQ', 'VG', 'VH', 'VI',
128
- 'VL', 'VK', 'VM', 'VF', 'VP', 'VS', 'VT', 'VW', 'VY', 'VV'
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
129
  ]
130
 
131
 
 
70
  # and is intentionally excluded from the input features.
71
  # ---------------------------------------------------------------------------
72
  selected_features = [
73
+ "_PolarizabilityC1", "_PolarizabilityC2", "_PolarizabilityC3",
74
+ "_SolventAccessibilityC1", "_SolventAccessibilityC2", "_SolventAccessibilityC3",
75
+ "_SecondaryStrC1", "_SecondaryStrC2", "_SecondaryStrC3",
76
+ "_ChargeC1", "_ChargeC2", "_ChargeC3",
77
+ "_PolarityC1", "_PolarityC2", "_PolarityC3",
78
+ "_NormalizedVDWVC1", "_NormalizedVDWVC2", "_NormalizedVDWVC3",
79
+ "_HydrophobicityC1", "_HydrophobicityC2", "_HydrophobicityC3",
80
+
81
+ "_PolarizabilityT12", "_PolarizabilityT13", "_PolarizabilityT23",
82
+ "_SolventAccessibilityT12", "_SolventAccessibilityT13", "_SolventAccessibilityT23",
83
+ "_SecondaryStrT12", "_SecondaryStrT13", "_SecondaryStrT23",
84
+ "_ChargeT12", "_ChargeT13", "_ChargeT23",
85
+ "_PolarityT12", "_PolarityT13", "_PolarityT23",
86
+ "_NormalizedVDWVT12", "_NormalizedVDWVT13", "_NormalizedVDWVT23",
87
+ "_HydrophobicityT12", "_HydrophobicityT13", "_HydrophobicityT23",
88
+
89
+ "_PolarizabilityD1001", "_PolarizabilityD1025", "_PolarizabilityD1050",
90
+ "_PolarizabilityD1075", "_PolarizabilityD1100",
91
+ "_PolarizabilityD2001", "_PolarizabilityD2025", "_PolarizabilityD2050",
92
+ "_PolarizabilityD2075", "_PolarizabilityD2100",
93
+ "_PolarizabilityD3001", "_PolarizabilityD3025", "_PolarizabilityD3050",
94
+ "_PolarizabilityD3075", "_PolarizabilityD3100",
95
+
96
+ "_SolventAccessibilityD1001", "_SolventAccessibilityD1025",
97
+ "_SolventAccessibilityD1050", "_SolventAccessibilityD1075",
98
+ "_SolventAccessibilityD1100",
99
+ "_SolventAccessibilityD2001", "_SolventAccessibilityD2025",
100
+ "_SolventAccessibilityD2050", "_SolventAccessibilityD2075",
101
+ "_SolventAccessibilityD2100",
102
+ "_SolventAccessibilityD3001", "_SolventAccessibilityD3025",
103
+ "_SolventAccessibilityD3050", "_SolventAccessibilityD3075",
104
+ "_SolventAccessibilityD3100",
105
+
106
+ "_SecondaryStrD1001", "_SecondaryStrD1025", "_SecondaryStrD1050",
107
+ "_SecondaryStrD1075", "_SecondaryStrD1100",
108
+ "_SecondaryStrD2001", "_SecondaryStrD2025", "_SecondaryStrD2050",
109
+ "_SecondaryStrD2075", "_SecondaryStrD2100",
110
+ "_SecondaryStrD3001", "_SecondaryStrD3025", "_SecondaryStrD3050",
111
+ "_SecondaryStrD3075", "_SecondaryStrD3100",
112
+
113
+ "_ChargeD1001", "_ChargeD1025", "_ChargeD1050",
114
+ "_ChargeD1075", "_ChargeD1100",
115
+ "_ChargeD2001", "_ChargeD2025", "_ChargeD2050",
116
+ "_ChargeD2075",
117
+ "_ChargeD3001", "_ChargeD3025", "_ChargeD3050",
118
+ "_ChargeD3075", "_ChargeD3100",
119
+
120
+ "_PolarityD1001", "_PolarityD1025", "_PolarityD1050",
121
+ "_PolarityD1075", "_PolarityD1100",
122
+ "_PolarityD2001", "_PolarityD2025", "_PolarityD2050",
123
+ "_PolarityD2075", "_PolarityD2100",
124
+ "_PolarityD3001", "_PolarityD3025", "_PolarityD3050",
125
+ "_PolarityD3075", "_PolarityD3100",
126
+
127
+ "_NormalizedVDWVD1001", "_NormalizedVDWVD1025",
128
+ "_NormalizedVDWVD1050", "_NormalizedVDWVD1075",
129
+ "_NormalizedVDWVD1100",
130
+ "_NormalizedVDWVD2001", "_NormalizedVDWVD2025",
131
+ "_NormalizedVDWVD2050", "_NormalizedVDWVD2075",
132
+ "_NormalizedVDWVD2100",
133
+ "_NormalizedVDWVD3001", "_NormalizedVDWVD3025",
134
+ "_NormalizedVDWVD3050", "_NormalizedVDWVD3075",
135
+ "_NormalizedVDWVD3100",
136
+
137
+ "_HydrophobicityD1001", "_HydrophobicityD1025",
138
+ "_HydrophobicityD1050", "_HydrophobicityD1075",
139
+ "_HydrophobicityD1100",
140
+ "_HydrophobicityD2001", "_HydrophobicityD2025",
141
+ "_HydrophobicityD2050", "_HydrophobicityD2075",
142
+ "_HydrophobicityD2100",
143
+ "_HydrophobicityD3001", "_HydrophobicityD3025",
144
+ "_HydrophobicityD3050", "_HydrophobicityD3075",
145
+ "_HydrophobicityD3100",
146
+
147
+ "A", "R", "N", "D", "C", "E", "Q", "G", "H", "I",
148
+ "L", "K", "M", "F", "P", "S", "T", "W", "Y", "V",
149
+
150
+ "AR", "AD", "AQ", "AG", "AL", "AK", "AF", "AP", "AT", "AV",
151
+ "RA", "RC", "RE", "RG", "RI", "RL", "RS", "RT", "RV",
152
+ "NR", "NC", "NG", "NI", "NP", "NS", "NY", "NV",
153
+ "DR", "DN", "DC", "DE", "DG", "DF", "DS", "DT", "DY",
154
+ "CR", "CN", "CD", "CC", "CI", "CL", "CK", "CT", "CY", "CV",
155
+ "EA", "ER", "ED", "EC", "EE", "EG", "EI", "EL", "EK",
156
+ "EF", "EP", "ET", "EV",
157
+ "QN", "QF", "QV",
158
+ "GA", "GR", "GC", "GE", "GG", "GI", "GL", "GK", "GF", "GP", "GY",
159
+ "HA", "HP", "HT",
160
+ "IA", "IR", "ID", "II", "IL", "IF", "IP", "IS", "IV",
161
+ "LA", "LR", "LD", "LC", "LG", "LI", "LK", "LM", "LF",
162
+ "LS", "LT", "LY", "LV",
163
+ "KA", "KN", "KC", "KG", "KI", "KL", "KK", "KP", "KY",
164
+ "MA", "MD", "ME", "MI", "MK", "MF", "MP", "MS", "MV",
165
+ "FR", "FE", "FQ", "FG", "FL", "FF", "FS", "FT", "FY", "FV",
166
+ "PA", "PR", "PC", "PE", "PL", "PK", "PP", "PS", "PV",
167
+ "SA", "SR", "SD", "SC", "SG", "SH", "SI", "SL", "SP", "ST", "SY",
168
+ "TA", "TR", "TC", "TE", "TQ", "TG", "TI", "TL", "TP", "TS", "TV",
169
+ "WA",
170
+ "YN", "YD", "YC", "YQ", "YG", "YP",
171
+ "VA", "VR", "VD", "VC", "VE", "VG", "VI", "VL", "VK",
172
+ "VS", "VT", "VY", "VV"
173
  ]
174
 
175