BinEX

Binning tool for improve binning percentage in existing binning methods

Run the script

run the python script with taxonomic label data

$python mahalanobis_binning.py 10s binned_contigs_features.csv unbinned_contigs_features.csv taxon.csv final_ouput.csv --p 1500 9



argument 1 : dataset name (ex : '10s')

argument 2 : initial binning method feature file (ex : binned_contigs_features.csv)

argument 3 : unbinned contigs feature file (ex : unbinned_contigs_features.csv)

argument 4 : taxon file [this contain the taxons of all contigs] (ex : taxon.csv )[OPTIONAL]

argument 5 : final output file (ex :final_ouput.csv)

argument 6 : indicate parameter selcting (ex: "--p") [OPTIONAL]

argument 7 : filtering length (ex: 1500) [OPTIONAL]

argument 8 : Critical value (ex: 9) [OPTIONAL]

sample dataset in shown in the test folder.

feature files for unbinned contigs and binned contigs should follow following structure.

Binned feature file headers => [contig_id, contig_length, bin , feature_1, feature_2,....., feature_N ]

Unbinned feature file headers => [contig_id, contig_length, feature_1, feature_2,....., feature_N ]

taxon file contian following structure (not a requirement to have headers in csv file)
```
Taxon file headers => [contig_id, Actual taxon]
```
argument 4 (taxon file) is needed to calculate accuracy and other properties of results. Even without that it is possible to run the script and get the output.
```
$python mahalanobis_binning.py 10s binned_contigs_features.csv unbinned_contigs_features.csv final_ouput.csv
```

You can pick the filtering length and another threshold value in command line option. Depending on those values the results may changes.The order of argument follow the following.

 --p [filtering length] [critical value]

eg :      
 taxon file available case :
      python mahalanobis_binning.py 10s sample_data/10s/10s_binned_contigs_features.csv  sample_data/10s/10s_unbinned_contigs_features.csv sample_data/10s/10s_taxon.csv sample_data/10s/10s_final_ouput.csv --p 1500 9
      
taxon file not available case : 
 python mahalanobis_binning.py 10s sample_data/10s/10s_binned_contigs_features.csv  sample_data/10s/10s_unbinned_contigs_features.csv sample_data/10s/10s_final_ouput.csv --p 1500 9

you can provide the relative paths or absolute path of files.
final output will display following format csv file
```
output file => [id, bin]
```
PythonTestBed.java file is there to run the script using java.(usefull for web applications). If it doesn't need (for normal run), make
```
testBedEnable = 0
```
for degugging perposes it is possible to enbale progressbar
```
progressBarEnable = 1
```

kasunvimu8/BinEX

BinEX

Run the script