#!/bin/env python
import argparse,argcomplete
import glob
import pickle as pkl
import _gdac_assembler as GDAS
import _gdac_analysis as GDAN
exec(open('/home/ramn/workdir/NGS-pipeline/scripts/HEADER/python-header.py').read())
exec(open('/home/ramn/workdir/NGS-pipeline/scripts/src/plotter.py').read())
exec(open('/home/ramn/workdir/NGS-pipeline/scripts/src/bookkeeping.py').read())

parser=argparse.ArgumentParser(description="assemble GDAC data")
parser.add_argument('--sample',help="--sample default=all",nargs='+',default=[])
parser.add_argument('--expression',help="--expression <expression filename> default=geneexpression_RSEM.txt",default="RNA-seq/geneexpression_RSEM.txt")
parser.add_argument('--clinical',help="--clinical <clinical filename> default=geneexpression_RSEM.txt",default="CLINICAL/clinical_merged.txt")
parser.add_argument('--annotation',help="--annotation <annotation filename> default=./annotation.gtf",default="./annotation.gtf")
parser.add_argument('--features',help="--features <featuresfile> default=./Haldar_rich_class_annotation.csv",default="./Haldar_rich_class_annotation.csv")
parser.add_argument('--plotgenes',nargs='+',help="--plotgenes <list of genes to plot> default= all in Haldar_rich_class",default=[])
parser.add_argument('--nosurvival',help="--nosurvival switches off survival plots",action='store_false')
argcomplete.autocomplete(parser)
args=parser.parse_args()

if len(args.sample)==0: args.sample=[_xx.split('/CLINICAL')[0] for _xx in glob.glob('../*/CLINICAL')]
_tcga = pd.DataFrame(columns=['cancer','group','Symbol','variable','value'])

for _sample in args.sample:
        print (_sample)
        outpklname = _sample+'/assembled_data.pkl'
        if not os.path.isfile(outpklname):
                print ('Processing sample:',_sample)
                _temp = GDAS.assembler(folder=_sample,rnaseqfile=args.expression,annotationfile=args.annotation, featurefile=args.features,clinicalfile=args.clinical)
                pkl.dump(_temp.datadict,open(outpklname,'wb'))
                print ('Completed analysis for:',_sample,'\n\n')
        else:
                _temp = pkl.load(open(outpklname,'rb'))

        for _group in ['normal','tumor']:
                if len(_temp['group'][_group])>0:
                        _temp1 = _temp['rna'][['Symbol']+_temp['group'][_group]]
                        _temp1['cancer'] = _sample.split('../')[-1]
                        _temp1['group'] = _group
                        _tcga = _tcga.append(_temp1,sort=False)

_tcga.to_json('GDAC_master_dataframe.json')