ECMPride

What is it?

ECMPride is a flexible and scalable tool developed for predicting extracellular matrix (ECM) proteins. ECMPride can directly perform ECM prediction by taking UniProt IDs in CSV (*.csv) file format as input. The core of ECMPride was written in R 3.6.1 language on the RStudio 1.1.442 under Windows System. The function in ECMPride are based on R statistical environment. Both single-threaded and multi-threaded versions of ECMPride are provided here.

The release version

ECMPride version v 1.3.0 (2/12/2020)

• Optimize the algorithm of ECMPride to reduce execution time;
• Provide the multi-thread version of ECMPride, which run faster when predicting large amount of proteins.

ECMPride version v 1.2.0 (1/3/2020)

• Update the positive dataset
  • We introduced more ECMs validated by the proteomic experiment of the healthy samples
  • We excluded some ECMs which appeared only in disease samples and don't have GO annotation of ECM.
  • The total number of ECMs is changed from 478 to 521.
• Update the prediction model.

ECMPride version v 1.1.0 (12/4/2019)

• Predictive model optimization: 99 submodels were used to solve the problem of unbalanced data sets;
• Enable prediction results be reproduced accurately by setting random number seeds before training models by random forest;
• Upload user manual to make it easier and clearer for users to use ECMPride;
• Fixed other bugs.

ECMPride version v 1.0.0 (3/11/2019)

• The first release version of ECMPride.

Hardware requirements

• 2.0 GHz CPU minimum
• 2 GB RAM minimum

Software requirements

• Supported operating system (OS) versions (32-bit or 64-bit) Windows 7 Windows 10

• R 3.6.1 or higher (for Windows) from R project

Note

The required R packages and their installation commands are listed below:

1. install.packages("randomForest")
2. install.packages("plyr")
3. install.packages("dplyr")
4. install.packages("xlsx")
5. install.packages("mRMRe")
6. install.packages("caret")
7. install.packages("parallel")

You should install these R packages by R 3.6.1 (not R 3.5.3 or the older version) ahead of time. In fact, ECMPride will install these R packages itself the first time it runs, but this approach may face some unknown errors. Therefore, we recommend users to install these R packages before running ECMPride.

Installation

Download link: https://github.com/Binghui-Liu/ECMPride.git.

Please see User Manual for ECMPride.pdf for details.

License

Please see the file called license.pdf.

Contact

For any questions involving ECMPride, please contact Mr. Binghui Liu (Email: l_binghui@163.com)