Datasets:

ArXiv:
License:
andrewfoong commited on
Commit
d2e9d05
1 Parent(s): 4030900

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +44 -1
README.md CHANGED
@@ -6,4 +6,47 @@ license: mit
6
 
7
  This dataset contains molecular dynamics simulation data that was used to train the neural networks in the NeurIPS 2023 paper [Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics](https://arxiv.org/abs/2302.01170).
8
 
9
- This dataset consists of many molecular dynamics trajectories of small peptides (2-4 amino acids) simulated with an implicit water force field.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
 
7
  This dataset contains molecular dynamics simulation data that was used to train the neural networks in the NeurIPS 2023 paper [Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics](https://arxiv.org/abs/2302.01170).
8
 
9
+ This dataset consists of many molecular dynamics trajectories of small peptides (2-4 amino acids) simulated with an implicit water force field.
10
+ For each protein two files are available:
11
+
12
+ * `protein-state0.pdb`: contains the topology and initial 3D XYZ coordinates.
13
+ * `protein-arrays.npz`: contains trajectory information.
14
+
15
+ The datasets are are split into the following directories:
16
+
17
+ # 2AA-1-big "Two Amino Acid" data set
18
+
19
+ This folder contains a data set of all-atom molecular dynamics trajectories for 380
20
+ of the 400 dipeptides, i.e. small proteins composed of two amino acids.
21
+ This dataset was orginally created missing 20 of the 400 possible dipeptides.
22
+ The `2AA-1-complete` dataset completes this by including all 400.
23
+ Each peptide is simulated using classical molecular dynamics and the
24
+ water is simulated using an implicit water model.
25
+ The trajectories are only saved every 10000 MD steps. There is no intermediate
26
+ spacing as for the other datasets for the Timewarp project.
27
+
28
+ # 2AA-1-complete "Two Amino Acid" data set
29
+
30
+ This folder contains a data set of all-atom molecular dynamics trajectories for all 400
31
+ dipeptides, i.e. small proteins composed of two amino acids.
32
+ This includes also the peptides missing in the other 2AA datasets.
33
+ Each peptide is simulated using classical molecular dynamics and the
34
+ water is simulated using an implicit water model.
35
+
36
+ # 4AA-huge "Four Amino Acid" data set, tetrapeptides
37
+
38
+ This folder contains a data set of all-atom molecular dynamics trajectories for
39
+ tetrapeptides, i.e. small proteins composed of four amino acids.
40
+ The data set contains mostly validation and test trajectories as it was mostly
41
+ used to validation and test purposes.
42
+ The training trajectories used are usually shorter.
43
+ Each peptide is simulated for 1 micro second using classical molecular dynamics and the
44
+ water is simulated using an implicit water model.
45
+
46
+ # 4AA-large "Four Amino Acid" data set, tetrapeptides
47
+
48
+ This folder contains a data set of all-atom molecular dynamics trajectories for
49
+ 2333 tetrapeptides, i.e. small proteins composed of four amino acids.
50
+ The data set is split into 1500 tetra-peptides in the train set, 400 in validation, and 433 in test.
51
+ Each peptide in the train set is simulated for 50ns using classical molecular dynamics and the
52
+ water is simulated using an implicit water model. Each other peptide is simulated for 500ns.