The dataset used by PPVED is as follows:
| Name | Source | Amount | Download | Description |
| total | UniProt+PubMed+Simulation | 12,865 |
here | set of benchmark and independent dataset |
| benchmark | UniProt+PubMed+Simulation | 10,292 | here | train and verify the model by stratified 5-fold CV |
| independent | UniProt+PubMed+Simulation | 2,573 | here | test the generalization ability of the model |
| newest | UniProt | 314 |
here | the newest dataset to further test the model |
Note: UniProt in 'total', 'benchmark', and 'independent' was collected on December 18, 2019; however, UniProt in 'newest' was collected on December 18, 2020.
2. Installation package
If you have numerous amino acid substitutions that need to be predicted, we strongly recommend that you download and use the local installation package of PPVED.
Applicable platform:
64-bit Linux
| Name | Version | Size | Download |
| PPVED | 1.0 | 4.2Gb | here |
Additionally, you can read the file readme in the installation package to learn how to install PPVED.
3. Dependent software
To successfully run PPVED, you must install the following seven software in advance. The installation of each software is easy, because readme gives you detailed instructions on how to install it.
| Name | Version | Size | Download from the original page | Download from our website | Installation instructions |
| SCRATCH | 1.2 | 6.3Gb | here | here | readme |
| EFICAz | 2.5.1 | 4.2Gb | here | here | readme |
| DIpro | 2.0 | 28Mb | here | here | readme |
| NetPhos | 3.1 | 700Kb | here | here | readme |
| UbPred | - | 1004Kb | here | here | readme |
| LocTree3 | - | 48Mb | here | here | readme |
| SOLpro | - | 12Mb | here | here | readme |
Note:
1. PPVED is absolutely academic, and does not involve any commercial interests.
2. All the links provided here are just for the convenience of users to download.