parse_medi.py

import os, cPickle, parse_drugbank


def main():
    name = "Mesna" # Acarbose
    #mapping_file = "/home/emre/data/indication/medi/MEDI_01212013_UMLS.csv"
    mapping_file = "/home/emre/data/indication/medi/MEDI_01212013.csv" 
    #name_to_cui_and_confidences = get_medi_mapping_old(mapping_file)
    name_to_icd_and_confidences = get_medi_mapping(mapping_file)
    print name_to_icd_and_confidences[name]
    return 


def get_disease_specific_drugs(drug_to_diseases, phenotype_to_mesh_id, cutoff=1):
    disease_to_drugs = {}
    mesh_id_to_phenotype = {}
    for phenotype, mesh_id in phenotype_to_mesh_id.items():
        mesh_id_to_phenotype[mesh_id] = phenotype
    for drugbank_id, diseases in drug_to_diseases.iteritems():
        for phenotype, dui, val in diseases:
            if val >= cutoff:
                if dui in mesh_id_to_phenotype: # In the disease data set
                    disease = mesh_id_to_phenotype[dui].lower()
                    disease_to_drugs.setdefault(disease, set()).add(drugbank_id)
    return disease_to_drugs


def get_drug_disease_mapping(name_to_icd_and_confidences, name_to_drug, synonym_to_drug, icd_to_mesh_ids, mesh_id_to_name, dump_file):
    """
    name_to_icd_and_confidences mapping from MEDI
    name_to_drug & synonym mapping from DrugBank
    icd_to_mesh_ids mapping from DO
    mesh_id_to_name mapping from UMLS (MSH | MH)
    """
    if dump_file is not None and os.path.exists(dump_file):
	drug_to_diseases = cPickle.load(open(dump_file))
	return drug_to_diseases 
    drug_to_diseases = {} # (mesh_term, mesh_id, association score) 
    for name, values in name_to_icd_and_confidences.iteritems():
	#if name != "Acarbose":
	#    continue
        # Get drugbank id from name in the label
	drugbank_id, drugbank_name = parse_drugbank.get_drugbank_id_from_name(name, name_to_drug, synonym_to_drug)
        if drugbank_id is None:
            continue
	#if drugbank_id != "DB00284":
	#    continue
	#print "%s\t%s\t%s" % (drugbank_name, drugbank_id, name)
        for icd, val in values:
	    icd = icd.split(".")[0]
	    #print icd, val, icd in icd_to_mesh_ids 
            if icd in icd_to_mesh_ids:
                dui_list = icd_to_mesh_ids[icd]
		#print dui_list
		for dui in dui_list:
		    if dui not in mesh_id_to_name:
			continue
		    phenotype = mesh_id_to_name[dui].lower()
		    #print dui, phenotype
		    drug_to_diseases.setdefault(drugbank_id, set()).add((phenotype, dui, val))
	    #else:
		# ICD9 ids are problematic - consider name matching
    if dump_file is not None:
	cPickle.dump(drug_to_diseases, open(dump_file, 'w'))
    return drug_to_diseases


def get_drug_disease_mapping_from_cui(name_to_cui_and_confidences, name_to_drug, synonym_to_drug, concept_id_to_mesh_id, mesh_id_to_name, dump_file):
    """
    name_to_cui_and_confidences mapping from MEDI
    name_to_drug & synonym mapping from DrugBank
    concept & mesh_id mapping from UMLS (MSH | MH)
    """
    if dump_file is not None and os.path.exists(dump_file):
	drug_to_diseases = cPickle.load(open(dump_file))
	return drug_to_diseases 
    drug_to_diseases = {} # (mesh_term, mesh_id, association score) 
    for name, values in name_to_cui_and_confidences.iteritems():
        # Get drugbank id from name in the label
	drugbank_id, drugbank_name = parse_drugbank.get_drugbank_id_from_name(name, name_to_drug, synonym_to_drug)
        if drugbank_id is None:
            continue
	print "%s\t%s\t%s" % (drugbank_name, drugbank_id, name)
        for cui, val in values:
	    #print cui, val, cui in concept_id_to_mesh_id 
            if cui in concept_id_to_mesh_id:
                dui = concept_id_to_mesh_id[cui]
                phenotype = mesh_id_to_name[dui]
                drug_to_diseases.setdefault(drugbank_id, set()).add((phenotype, dui, val))
    if dump_file is not None:
	cPickle.dump(drug_to_diseases, open(dump_file, 'w'))
    return drug_to_diseases


def get_medi_mapping(mapping_file, textual_indication=False):
    """ 
    RXCUI_IN,DRUG_DESC,ICD9,INDICATION_DESCRIPTION,MENTIONEDBYRESOURCES,HIGHPRECISIONSUBSET,POSSIBLE_LABEL_USE
    44,Mesna,599.7,Hematuria,1,0,0
    """
    name_to_icd_and_confidences = {}
    f = open(mapping_file)
    header = f.readline()
    for line in f:
        words = line.strip().split(",")
        name = words[1]
        icd = words[2]
	if textual_indication:
	    icd = words[3]
        in_hps = words[5]
        confidence = 0.5
        if in_hps == "1":
            confidence = 1 
        name_to_icd_and_confidences.setdefault(name, []).append((icd, confidence))
    f.close()
    return name_to_icd_and_confidences
 

def get_medi_mapping_umls(mapping_file):
    """ 
    RXCUI_IN,STR,CUI,ATC,NDDF,CODE,HSP
    44,mesna,C0000294,R05CB05,2678,595,1
    """
    name_to_icd_and_confidences = {}
    f = open(mapping_file)
    header = f.readline()
    for line in f:
        words = line.strip().split(",")
        name = words[1]
        icd = words[-2]
        in_hps = words[-1]
        confidence = 0.5
        if in_hps == "1":
            confidence = 1 
        name_to_icd_and_confidences.setdefault(name, []).append((icd, confidence))
    f.close()
    return name_to_icd_and_confidences
    

def get_medi_mapping_old(mapping_file, textual_indication=False):
    """ 
    RXCUI_IN,DRUG_DESC,UMLS_CUI,ICD9,INDICATION_DESCRIPTION,MENTIONEDBYRESOURCES,HIGHPRECISIONSUBSET,POSSIBLE_LABEL_USE
    44,Mesna,C0018965,599.7,Hematuria,1,0,0
    """
    name_to_cui_and_confidences = {}
    f = open(mapping_file)
    header = f.readline()
    for line in f:
        words = line.strip().split(",")
        name = words[1]
        cui = words[2]
	if textual_indication:
	    cui = words[3]
        in_hps = words[6]
        confidence = 0.5
        if in_hps == "1":
            confidence = 1 
        name_to_cui_and_confidences.setdefault(name, []).append((cui, confidence))
    f.close()
    return name_to_cui_and_confidences
    
	
if __name__ == "__main__":
    main()