ML-locMLFE is a novel prediction multi-label protein subcellular localization model, which is able to use sparse reconstruction of the training samples to represent the bottom layer of the feature space.

ML-locMLFE is developed under Windows environment with:
matlab 2015b python 3.6.1
numpy 1.16.1
pandas 0.20.1

1.Build newPlant dataset following (https://www.uniprot.org/uniprot/) 2.Build SARS-CoV-2 dataset following (https://www.covid19dataportal.org/)

1. The dataset file contains Gram-positive bacteria dataset, newplant dataset, SARS-CoV-2 dataset, virus dataset and Gram-negative bacteria dataset.
2. Feature extraction: mainEBGW.m, ebgw1.m, ebgw2.m, ebgw3.m is the implementation of EBGW. PAAC.m,mainpseaac.m is the implementation of PseAAC. EDT.py is the implementation of EDT. RPT.py is the implementation of RPT. MCD.m, MCD1D.m, MCD2D.m, MCD3D.m, MCDexchange.m, MCDfeature.m, MCDfen.m, MCDtransform.m, MCDZD.m is the implementation of MCD. Gene Ontology can be found from http://www.ebi.ac.uk/GOA/.
3. Dimensional reduction: MLSI_transform.m represents MLSI. MDDM_transform.m represents MDDM. MVMD_transform.m represents MVMD. PCA_transform.m represents PCA. GRRO.m represents GRRO. MDFS.m represents MDFS.
4. Classifier: MLFE_train.m, MLFE_test is the implementation of MLFE. LIFT.m is the implementation of LIFT. MLKNN_train.m, MLKNN_test.m are the implementation of MLKNN. ML_RBF_train.m, ML_RBF_test.m is the implementation of ML_RBF. RankSVM_train.m, RankSVM_test.m is the implementation of RankSVM. InsDif.m is the implementation of InsDif.
5. Model: An example is included in the Demo file, and you can run the demo.m in MATLAB.

Contact: Bin Yu ([email protected])