update

Author: atante

README.md

@@ -2,8 +2,6 @@
 
 _The above picture is exported from_ [Wikipedia](https://en.wikipedia.org/wiki/Metabolic_pathway).
 
-## mlLGPR is currently not maintained. A new model is coming soon. Please stay tuned.
-
 ## Basic Description
 
 mlLGPR (**m**ulti-**l**abel **L**ogistic Re**G**ression for **P**athway P**R**ediction) is a novel pathway prediction framework to recover metabolic pathways from large-scale metagenomics datasets and tackle some pathway related obstacles. Unlike the conventional supervised methods that assume each sample is associated with a single class label within a number of candidate classes, a metagenomics dataset usually comprises of multiple pathways per sample, thus, putting the problem in the context of a multi-label classification approach. The originality of our method lies:
@@ -28,12 +26,18 @@ Our comprehensive analysis using seven designed experimental protocols demonstra
 To display mlLGPR's running options, use: `python main.py --help`. It should be self-contained. 
 
 ### Basic Usage
-All the command arguments are initiated through [main.py](main.py) file. You need to obtain [MetaCyc](https://metacyc.org/) database in order to extract information. Please modify the content of ``Path.py`` inside utility folder as necessary.
+All the command arguments are initiated through [main.py](main.py) file. 
+- You need to obtain [MetaCyc](https://metacyc.org/) database in order to extract information. Please modify the content of ``Path.py`` inside utility folder as necessary.
+- In addition, please download six database: AraCyc, EcoCyc, HumanCyc, LeishCyc, TrypanoCyc, and YeastCyc from [biocyc](https://biocyc.org/).
+
 
 #### Example
 To extract information from [MetaCyc](https://metacyc.org/), create golden and synthetic samples, train mlLGPR using elastic-net, evaluate, and predict on dataset, simply set the arguments in the [main.py](main.py) file as:
 
-```python main.py --biocyc --metagenomic --train --evaluate --predict --build_syn_dataset --nSample 15000 --average_item_per_sample 500 --build_synthetic_features --build_golden_dataset --build_golden_features --extract_info_mg --build_mg_features --ds_type "syn_ds" --trained_model "mlLGPR_en_ab_re_pe.pkl" --n_jobs 10 --nEpochs 10 --nBatches 5```
+```python main.py --biocyc --train --evaluate --predict --build_syn_dataset --nSample 15000 --average_item_per_sample 500 --build_synthetic_features --build_golden_dataset --build_golden_features --extract_info_mg --build_mg_features --ds_type "syn_ds" --trained_model "mlLGPR_en_ab_re_pe.pkl" --kbpath "[MetaCyc location]" --dspath "[Location to the processed dataset]" --mdpath "[Location to store or save the model]" --rspath "[Resuls location]" --ospath "[Object location]" --n_jobs 10 --nEpochs 10 --nBatches 5```
+
+where 
+- Object location: The location to the data object that contains extracted information from the MetaCyc database and all the datbases.
 
 
 ## Citing