about the initial features of protein
Closed this issue · 2 comments
SakuraRiven commented
Hi，thanks for sharing your work!
As a novice, I am curious about the initial features of protein. Besides the one-hot of amino acid, how can we obtain the other features? Is there any reference？
Thanks for your help!~
yjh126yjh commented
Each of the last 17 dimensions represents a chemical/physical property of amino acid. Here is the reference: Table of standard amino acid abbreviations and properties
The 38 features are:
- 0: G
- 1: P
- 2: T
- 3: E
- 4: S
- 5: K
- 6: C
- 7: L
- 8: M
- 9: V
- 10: D
- 11: A
- 12: R
- 13: I
- 14: N
- 15: H
- 16: F
- 17: W
- 18: Y
- 19: Q
- 20: X
- 21: Side Chain Class is acid
- 22: Side Chain Class is aliphatic
- 23: Side Chain Class is amide
- 24: Side Chain Class is aromatic
- 25: Side Chain Class is basic
- 26: Side Chain Class is cyclic
- 27: Side Chain Class is hydroxyl-containing
- 28: Side Chain Class is sulfur-containing
- 29: Side Chain Polarity is acidic polar
- 30: Side Chain Polarity is nonpolar
- 31: Side Chain Polarity is polar
- 32: Side Chain Polarity is basic polar
- 32: Side Chain Polarity is basic polar
- 33: Side Chain Charge is negative
- 34: Side Chain Charge is neutral
- 35: Side Chain Charge is positive
- 36: Hydropathy Index (non onehot. Original value is divided by 5)
- 37: Molecular Mass（non onehot. Original value is subtracted by mean then divided by 70)
SakuraRiven commented
Understand. Thanks again for your answer!