@@ -13,52 +13,39 @@ class PSeAAC:
1313 selected physicochemical properties and sequence-order correlations as described in
1414 the PseAAC model by Chou.
1515
16- The PSeAAC algorithm uses normalized physicochemical (NP) properties of amino
17- acids, which we load from a predefined matrix using `aa_props`. These properties
18- can be grouped using one of the following options:
19-
20-
21- - `prop_indices`: A list of property indices (0-based) to use (e.g.,
22- `prop_indices=[3, 4, 8]` uses property groups [Polarity, Molecular Weight,
23- Bulkiness] (refer table below)).
24- - If `group_props` is an integer, the selected properties are grouped into chunks of
25- that size (e.g., `group_props=3` groups into sets of 3). If `group_props` is None,
26- then each property is treated as its own group.
27- - If `custom_groups` is provided, it overrides all other logic. It must be a list of
28- lists, where each sublist contains indices (0-based) of properties to group
29- together. (e.g., `prop_indices=[[3, 4], [8]]` groups Polarity and Molecular Weight
30- together, and Bulkiness is its own group).
16+ The PSeAAC algorith uses 21 normalized physiochemical (NP) properties of amino
17+ acids, which we load from a predefined matrix using `aa_props`.These 21 properties
18+ are grouped into 7 distinct property groups, with each group containing
19+ 3 consecutive properties. Specifically, the groups are arranged in order as follows:
20+ Group 1 includes properties 1–3, Group 2 includes properties 4–6, and so on, up to
21+ Group 7, which includes properties 19–21.
3122
3223
3324 The properties in order are:
34-
35-
36- 0. Hydrophobicity
37- 1. Hydrophilicity
38- 2. Side-chain Mass
39- 3. Polarity
40- 4. Molecular Weight
41- 5. Melting Point
42- 6. Transfer Free Energy
43- 7. Buriability
44- 8. Bulkiness
45- 9. Solvation Free Energy
46- 10. Relative Mutability
47- 11. Residue Volume
48- 12. Volume
49- 13. Amino Acid Distribution
50- 14. Hydration Number
51- 15. Isoelectric Point
52- 16. Compressibility
53- 17. Chromatographic Index
54- 18. Unfolding Entropy Change
55- 19. Unfolding Enthalpy Change
56- 20. Unfolding Gibbs Free Energy Change
25+ 1. Hydrophobicity
26+ 2. Hydrophilicity
27+ 3. Side-chain Mass
28+ 4. Polarity
29+ 5. Molecular Weight
30+ 6. Melting Point
31+ 7. Transfer Free Energy
32+ 8. Buriability
33+ 9. Bulkiness
34+ 10. Solvation Free Energy
35+ 11. Relative Mutability
36+ 12. Residue Volume
37+ 13. Volume
38+ 14. Amino Acid Distribution
39+ 15. Hydration Number
40+ 16. Isoelectric Point
41+ 17. Compressibility
42+ 18. Chromatographic Index
43+ 19. Unfolding Entropy Change
44+ 20. Unfolding Enthalpy Change
45+ 21. Unfolding Gibbs Free Energy Change
5746
5847
5948 Each feature vector consists of:
60-
61-
6249 - 20 normalized amino acid composition features (frequency of each standard
6350 amino acid)
6451 - 30 sequence-order correlation features based on physicochemical similarity between
@@ -67,7 +54,6 @@ class PSeAAC:
6754 resulting in a final vector of length 50 × number of property groups.
6855 For example, if there are 7 groups, the output length is 350 ((20 + 30) * 7 = 350).
6956
70-
7157 References
7258 ----------
7359 Shen HB, Chou KC. PseAAC: a flexible web server for generating various kinds of
0 commit comments