bug修复

tools/parsexlsx.py

@@ -19,7 +19,7 @@ class BaseAssignment:
         self.result = tree()  # 报告结果
         self.signtb = set()  # 具有明确或潜在临床意义的基因变异
         self.signdurg = set()  # 潜在临床获益的治疗药物
-
+        self.sign_from_type = defaultdict(dict)  # 不同来源有意义突变记录，NCCN使用
         self.drugs_type = dict()
 
 
@@ -37,7 +37,8 @@ class Parse(BaseAssignment):
         if data.empty:
             raise UserWarning('sample_info表为空，生成报告失败！')
         data = data.applymap(
-            lambda x: str(x).replace('.', '/').replace('-', '/').replace('——', '/') if str(x) in ['.', '-', '——'] else x)
+            lambda x: str(x).replace('.', '/').replace('-', '/').replace('——', '/') if str(x) in ['.', '-',
+                                                                                                  '——'] else x)
         data_dict = data.to_dict('index')[0]
         data_dict['receiveTime'] = re.split(' ', data_dict['receiveTime'])[0]
         data_dict['reportTime'] = time.strftime("%Y-%m-%d", time.localtime())
@@ -60,7 +61,8 @@ class Parse(BaseAssignment):
         # 拆分hgvs
         data[['gene', 'transcript', 'exon', 'nacid', 'aacid']] = data['AAChange.refGene'].str.split(':', expand=True)
         # 没有氨基酸改变用核苷酸改变代替
-        data['aacid'] = data['aacid'].fillna(data['nacid'])
+        # data['aacid'] = data['aacid'].fillna(data['nacid'])
+        data['aacid'] = data['aacid'].fillna('/')
 
         for alter, alter_data in data.groupby('AAChange.refGene'):
             alter_data_need = alter_data[['gene', 'transcript', 'exon', 'nacid', 'aacid', 'mutant_frequency',
@@ -77,6 +79,7 @@ class Parse(BaseAssignment):
             # 汇总
             if alter_res['AMP_mut_level'] in ['I', 'II']:
                 self.signtb.add(alter)
+                self.sign_from_type['target'][alter_res['gene']] = '%s %s' % (alter_res['gene'], alter_res['nacid'])
 
         self.result['snvindel'] = res
 
@@ -88,19 +91,21 @@ class Parse(BaseAssignment):
             return
         data = data[data['Validated'] == 1].reset_index()
         for alter, alter_data in data.groupby('FUSION'):
-            alter_data_need = alter_data[['FUSION', 'FREQ1', 'AMP_mut_level', 'Gene_function']]
+            alter_data_need = alter_data[['FUSION', 'FREQ1', 'AMP_mut_level', 'Gene_function', 'Gene_Symbol']]
             alter_res = alter_data_need.iloc[0].to_dict()
             alter_res['drug_category'] = self._drug_category(alter_data)
             drug_content = alter_data[
                 ['DrugCn', 'Response_Type', 'Indication', 'Evidence_Source', 'Efficacy_Evidence']]
             drug_content = drug_content[drug_content['DrugCn'] != '.']
             alter_res['drug_content'] = drug_content.reset_index().to_dict('records')
-            alter_res['alter'] = '%s 融合' % (alter_res['FUSION'].replace('-', ':'))
+            des = '%s 融合' % (alter_res['FUSION'].replace('-', '::'))
+            alter_res['alter'] = des
             res.append(alter_res)
 
             # 汇总
             if alter_res['AMP_mut_level'] in ['I', 'II']:
                 self.signtb.add(alter)
+                self.sign_from_type['fusion'][alter_res['Gene_Symbol']] = des
 
         self.result['fusion'] = res
 
@@ -121,12 +126,14 @@ class Parse(BaseAssignment):
                 ['DrugCn', 'Response_Type', 'Indication', 'Evidence_Source', 'Efficacy_Evidence']]
             drug_content = drug_content[drug_content['DrugCn'] != '.']
             alter_res['drug_content'] = drug_content.reset_index().to_dict('records')
+            des = '%s %s' % (alter, alter_res['muttype'])
             alter_res['alter'] = '%s %s' % (alter, alter_res['muttype'])
             res.append(alter_res)
 
             # 汇总
             if alter_res['AMP_mut_level'] in ['I', 'II']:
                 self.signtb.add(alter)
+                self.sign_from_type['cnv'][alter_res['Gene_Symbol']] = des
         self.result['cnv'] = res
 
     def hotspot(self):
@@ -268,13 +275,21 @@ class Parse(BaseAssignment):
 
     def chemo(self):
         chemo_res = self._to_records('chemo_res', need=True)
-        chemo_res_df = pd.DataFrame(chemo_res)
-        chemo_res_df.index = chemo_res_df.index + 1
-        chemo_res_df = chemo_res_df.reset_index()
-        self.result['chemo']['chemo_res'] = chemo_res_df.to_dict('records')
-        self.result['sum']['chemo']['drug_num'] = len(chemo_res)
-        self.result['sum']['chemo']['drug_category'] = pd.DataFrame(chemo_res).groupby('推荐程度')['药物名称'].apply(
-            ','.join).to_dict()
+
+        chemo_res_list = []
+        chemo_sign_drug_num = 0
+        chemo_drug_category = dict()
+        if chemo_res:
+            chemo_res_df = pd.DataFrame(chemo_res)
+            chemo_res_df.index = chemo_res_df.index + 1
+            chemo_res_df = chemo_res_df.reset_index()
+            chemo_res_list = chemo_res_df.to_dict('records')
+            chemo_sign_drug_num = len(chemo_res_df[chemo_res_df['推荐程度'] == '推荐'])
+            chemo_drug_category = chemo_res_df.groupby('推荐程度')['药物名称'].apply(
+                ','.join).to_dict()
+        self.result['chemo']['chemo_res'] = chemo_res_list
+        self.result['sum']['chemo']['drug_num'] = chemo_sign_drug_num
+        self.result['sum']['chemo']['drug_category'] = chemo_drug_category
 
         chemo_comb = self._to_records('chemo_comb', need=True)
         chemo_comb_res = dict()
@@ -309,13 +324,15 @@ class Parse(BaseAssignment):
 
         hereditary_risk = pd.DataFrame(self.sampledata['hereditary_risk'])
         if not hereditary_risk.empty:
-            risk = ','.join(hereditary_risk[hereditary_risk['风险值'] == '偏高']['肿瘤类型'].to_list())
+            risk = ','.join(hereditary_risk[hereditary_risk['风险值'] == '偏高']['肿瘤类型'].to_list()) + '风险可能较高'
 
         self.result['hereditary'] = hereditary.to_dict('records')
         self.result['sum']['hereditary']['result'] = result
         self.result['sum']['hereditary']['disease'] = disease
         self.result['sum']['hereditary']['risk'] = risk
 
+        self.result['sum']['hereditary']['num'] = len(hereditary.index)
+
     def qc(self):
         # self._to_dicts('qc')
         data = pd.DataFrame(self.sampledata['qc'])
@@ -335,11 +352,35 @@ class Parse(BaseAssignment):
         if not data.empty:
             data = data.dropna()
             data = data[data['drug_detail'] != '.']
+            data = data.applymap(lambda x: str(x).replace('\\\\', '\n') if ' \\\\' in str(x) else x)
             res = data.set_index('drug_name')['drug_detail'].to_dict()
         self.result['drugs']['drugs_detail'] = res
 
     def indication(self):
-        self._to_records('indication')
+        indication_res = self._to_records('indication', need=True)
+        trans = dict(
+            突变='target',
+            融合='fusion',
+            扩增='cnv'
+        )
+        res = list()
+        if indication_res:
+            # indication_res_df = pd.DataFrame(indication_res)
+            # indication_res_df['变异'] = indication_res_df.apply(
+            #     lambda x: self.sign_from_type.get(trans.get(tbtype, ''), '') for tbtype in x['检测内容'].split('/'))
+            for indication_sp in indication_res:
+                gene_tbtype_res = list()
+                for tbtype in indication_sp['检测内容'].split('/'):
+                    if tbtype not in trans:
+                        continue
+                    if trans[tbtype] not in self.sign_from_type:
+                        continue
+                    if indication_sp['基因'] not in self.sign_from_type[trans[tbtype]]:
+                        continue
+                    gene_tbtype_res.append(self.sign_from_type[trans[tbtype]][indication_sp['基因']])
+                indication_sp['检测情况'] = '\n'.join(gene_tbtype_res)
+                res.append(indication_sp)
+        self.result['indication'] = res
 
     def _to_records(self, sheetname, need=False):
         """
@@ -376,17 +417,21 @@ class Parse(BaseAssignment):
         for drug_category, drug_category_alter_data in groupdata.groupby('Drug_Category'):
             if drug_category == '.':
                 continue
+            drug_category_alter_data['drug_split'] = drug_category_alter_data['DrugCn'].str.split(',')
+            drug_category_alter_data_split = drug_category_alter_data.explode('drug_split').reset_index()
             # 敏感，可能敏感药物统计
             if drug_category in ['a', 'b', 'c']:
-                self.signdurg.update(set(drug_category_alter_data['DrugCn'].str.split(',').explode().tolist()))
-            drug_category_alter_data['drugdes'] = drug_category_alter_data.apply(
-                lambda x: '%s 【%s 级】' % (x['DrugCn'], x['AMP_evidence_level']), axis=1)
-            drug_category_res[drug_category] = '\n'.join(drug_category_alter_data['drugdes'].to_list())
+                self.signdurg.update(set(drug_category_alter_data_split['drug_split'].tolist()))
+            drug_category_alter_data_split['drugdes'] = drug_category_alter_data_split.apply(
+                lambda x: '%s 【%s 级】' % (x['drug_split'], x['AMP_evidence_level']), axis=1)
+            drug_category_res[drug_category] = '\n'.join(drug_category_alter_data_split['drugdes'].to_list())
 
         # 所有药物信息
         groupdata['list_col'] = groupdata['DrugCn'].str.replace(' + ', '+').str.split(r'[+,]')
         exploded_df = groupdata.explode('list_col').reset_index()
         exploded_df = exploded_df[(exploded_df['list_col'] != '.') & (exploded_df['list_col'] != '')]
+        exploded_df.loc[exploded_df['Response_Type'].str.contains('敏感'), 'Response_Type'] = '可能敏感'
+        exploded_df.loc[exploded_df['Response_Type'].str.contains('耐药'), 'Response_Type'] = '可能耐药'
         exploded_dict = exploded_df.groupby('Response_Type')['list_col'].agg(lambda x: list(set(x))).to_dict()
 
         for drug_type in exploded_dict:
@@ -448,8 +493,8 @@ def run(path):
     parse = Parse(read(path))
     res = parse.collect()
     resjson = json.dumps(res, indent=4, ensure_ascii=False)
-    with open('t.json', 'w') as f:
-        f.write(resjson)
+    # with open('t.json', 'w') as f:
+    #     f.write(resjson)
     return resjson