更新qcfail

codes/postprocess.py

@@ -91,6 +91,7 @@ class PostProcess:
         # 突变的等级分类
         if '证据等级' in df.columns:
             df['AMP_mut_level'] = df['证据等级'].replace(['A', 'B', 'C', 'D'], ['I', 'I', 'II', 'II'])
+            # df['NCCN_mut_level'] = df['证据等级'].replace(['A', 'B', 'C', 'D'], ['I', 'II', 'II', 'II'])
 
         # 药物等级划分
         if 'Response_Type_C' in df.columns and '证据等级' in df.columns:
@@ -527,7 +528,8 @@ class PostProcess:
         if qc_files:
             if os.path.getsize(qc_files[0]) > 0:
                 qc = pd.read_csv(qc_files[0], sep='\t', header=None)
-                qc_res = [qc.set_index(0)[1].to_dict()]
+                qc_content = qc.set_index(0)[1].to_dict()
+                qc_res = [qc_content]
         self.sheet['qc'] = qc_res
 
     def drugs(self):

codes/qc_check.py

@@ -0,0 +1,100 @@
+#!/usr/bin/env python3
+
+import argparse
+import json
+import os.path
+
+import pandas as pd
+
+
+def check(outputf, tumorf, normalf=None, genderf=None, cms=None):
+    # qc_check = {
+    #     'Q30(%)': 85,
+    #     'clean_reads_rate(%)': 90,
+    #     'mapped_rate(%)': 90,
+    #     'capture_rate(reads)': 50,
+    #     'mean_depth(dedup)': 500,
+    #     'mean_depth(dedup)_umi': 1000,
+    #     'coverage(>=80%)': 500,
+    #     'coverage(>=80%)_umi': 1000,
+    #     'coverage(>=0.2*meanx)': 95,
+    #     'coverage(>10x)': 90,
+    #     'coverage(>=0.1*meanx)': 90,
+    #     'concordence': 99,
+    #     'contamination_n': 1,
+    #     'contamination_t': 1,
+    # }
+
+    qc_t = pd.read_csv(tumorf, sep="\t", index_col=0, header=None).T
+
+    out = open(outputf, 'w')
+
+    ##tumor
+    if float(qc_t['Q30(%)'][1]) < 85:
+        out.write('\t'.join(("Tumor:", "Q30(%)", str(qc_t['Q30(%)'][1]), "(<85%)\n")))
+    if float(qc_t['clean_reads_rate(%)'][1]) < 90:
+        out.write('\t'.join(("Tumor:", "clean_reads_rate(%)", str(qc_t['clean_reads_rate(%)'][1]), "(<90%)\n")))
+    if float(qc_t['mapped_rate(%)'][1]) < 90:
+        out.write('\t'.join(("Tumor:", "mapped_rate(%)", str(qc_t['mapped_rate(%)'][1]), "(<90%)\n")))
+    if float(qc_t['capture_rate(reads)'][1]) < 50:
+        out.write('\t'.join(("Tumor:", "capture_rate(reads)", str(qc_t['capture_rate(reads)'][1]), "(<50%)\n")))
+    if float(qc_t['mean_depth(dedup)'][1]) < 500:
+        out.write('\t'.join(("Tumor:", "mean_depth(dedup)", str(qc_t['mean_depth(dedup)'][1]), "(<500X)\n")))
+    if float(qc_t['coverage(>=80%)'][1]) < 500:
+        out.write('\t'.join(("Tumor:", "coverage(>=80%)", str(qc_t['coverage(>=80%)'][1]), "(<500X)\n")))
+    if float(qc_t['coverage(>=0.2*meanx)'][1]) < 95:
+        out.write('\t'.join(("Tumor:", "coverage(>=0.2*meanx)", str(qc_t['coverage(>=0.2*meanx)'][1]), "(<95%)\n")))
+    if float(qc_t['coverage(>10x)'][1]) < 95:
+        out.write('\t'.join(("Tumor:", "coverage(>10x)", str(qc_t['coverage(>10x)'][1]), "(<95%)\n")))
+
+    if os.path.exists(normalf):
+        qc_n = pd.read_csv(normalf, sep="\t", index_col=0, header=None).T
+        ##normal
+        if float(qc_n['Q30(%)'][1]) < 85:
+            out.write('\t'.join(("Normal:", "Q30(%)", str(qc_n['Q30(%)'][1]), "(<85%)\n")))
+        if float(qc_n['clean_reads_rate(%)'][1]) < 90:
+            out.write('\t'.join(("Normal:", "clean_reads_rate(%)", str(qc_n['clean_reads_rate(%)'][1]), "(<90%)\n")))
+        if float(qc_n['mapped_rate(%)'][1]) < 90:
+            out.write('\t'.join(("Normal:", "mapped_rate(%)", str(qc_n['mapped_rate(%)'][1]), "(<90%)\n")))
+        if float(qc_n['capture_rate(reads)'][1]) < 50:
+            out.write('\t'.join(("Normal:", "capture_rate(reads)", str(qc_n['capture_rate(reads)'][1]), "(<50%)\n")))
+        if float(qc_n['coverage(>=80%)'][1]) < 100:
+            out.write('\t'.join(("Normal:", "coverage(>=80%)", str(qc_n['coverage(>=80%)'][1]), "(<100X)\n")))
+        if float(qc_n['coverage(>=0.2*meanx)'][1]) < 95:
+            out.write(
+                '\t'.join(("Normal:", "coverage(>=0.2*meanx)", str(qc_n['coverage(>=0.2*meanx)'][1]), "(<95%)\n")))
+        if float(qc_n['coverage(>10x)'][1]) < 95:
+            out.write('\t'.join(("Normal:", "coverage(>10x)", str(qc_n['coverage(>10x)'][1]), "(<95%)\n")))
+    if os.path.exists(str(genderf)):
+        genderf_df = pd.read_csv(genderf, sep='\t')
+        genderf_dict = genderf_df.to_dict(orient='records')[0]
+        gender_cms = ''
+        if os.path.exists(str(cms)):
+            cmsf = open(cms, "r")
+            cmsj = json.load(cmsf)
+            gender_cms_data = cmsj.get('data', [])
+            if gender_cms_data:
+                gender_cms = gender_cms_data[0].get('gender', '')
+        if gender_cms:
+            genderf_dict['gender_cms'] = gender_cms
+            genderf_r = [k for k, v in genderf_dict.items() if list(genderf_dict.values()).count(v) == 1]
+            if genderf_r:
+                out.write('\t'.join(("性别鉴定不一致:", str(genderf_dict))))
+        else:
+            out.write('\t'.join(("性别鉴定:", 'lims系统提供信息有误，请核查')))
+    else:
+        out.write('\t'.join(("性别鉴定:", '流程未进行性别鉴定评估')))
+
+    out.close()
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description="Check qcfile")
+    parser.add_argument('-n', '--barcode', help="sample's qcfile", required=True)
+    parser.add_argument('-s', '--normal', help="normal's file", default='', required=False, nargs='?')
+    parser.add_argument('-p', '--gender', help="gender file", default='', required=False, nargs='?')
+    parser.add_argument('-c', '--cms', help="cms's file", default='', required=False, nargs='?')
+    parser.add_argument('-o', '--output_file', help="Output directory, default ", required=True)
+
+    args = parser.parse_args()
+    check(args.output_file, args.barcode, args.normal, args.gender, args.cms)

codes/run_wdl.py

@@ -66,12 +66,13 @@ def get_catecode(project, start_node='addQc'):
     return cate_dict
 
 
-def send_ding(msg):
+def send_ding(msg, isat=True):
     '''
     发送钉钉消息功能
     '''
     url = 'https://oapi.dingtalk.com/robot/send?access_token=0c4b2dc1b6a1b459826512cc27adbd14e4f6aa2e661b7a7c284669065bbccfc5'
-    data = {"msgtype": "text", "text": {"content": "pipeline:" + str(msg)}, "at": {"isAtAll": True}}
+    # url = 'https://oapi.dingtalk.com/robot/send?access_token=d4a0749cc7ff87bd12079a79dd74ca3423becb1ce161c3088acc6628a7a188dd'
+    data = {"msgtype": "text", "text": {"content": "pipeline:" + str(msg)}, "at": {"isAtAll": isat}}
     headers = {'Content-Type': 'application/json;charset=UTF-8'}
     send_data = json.dumps(data).encode('utf-8')
     session.post(url=url, data=send_data, headers=headers)
@@ -116,7 +117,7 @@ def run(barcode, normal, umi, input_dir, output_dir, project, cancer, probe, wdl
     # 记录开始时间
     start_time = time.time()
     if not debug:
-        send_ding(msg=f'\n样本: {barcode}\n分析地址: \n{output_dir} \n 开始分析')
+        send_ding(msg=f'\n样本: {barcode}\n分析地址: \n{output_dir} \n 开始分析', isat=False)
     ret = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, encoding="utf-8")
     pidnum = ret.pid
     with open(os.path.join(output_dir, 'pid'), 'w') as pidfile:
@@ -163,12 +164,12 @@ def run(barcode, normal, umi, input_dir, output_dir, project, cancer, probe, wdl
     # 按照文件的修改时间倒序排序
     sorted_stdout_files = sorted(stdout_files, key=lambda x: os.path.getmtime(x), reverse=True)
     sorted_stderr_files = sorted(stderr_files, key=lambda x: os.path.getmtime(x), reverse=True)
-
     for file_path in sorted_stdout_files:
         with open(file_path, 'r') as f:
             stdout.write('#' * 50 + '\n')
             stdout.write(file_path + '\n')
-            stdout.write(f.read())
+            f_read = f.read()
+            stdout.write(f_read)
             stdout.write('\n\n')
 
     for file_path in sorted_stderr_files:
@@ -177,8 +178,18 @@ def run(barcode, normal, umi, input_dir, output_dir, project, cancer, probe, wdl
             stderr.write(file_path + '\n')
             stderr.write(f.read())
             stderr.write('\n\n')
+
+    qcfail_f = os.path.join(output_dir, 'report', f'{barcode}_qcfail.txt')
+    if os.path.exists(qcfail_f):
+        with open(qcfail_f, "r") as qcfail_f_out:
+            qcfail = qcfail_f_out.read()
+    else:
+        qcfail = '流程失败，请核查！！！'
+    if not qcfail:
+        qcfail = '合格'
     if not debug:
-        send_ding(msg=f'\n样本: {barcode}\n分析地址: \n{output_dir} \n 分析完成，请查看！')
+        send_ding(msg=f'\n样本: {barcode}\n分析地址: \n{output_dir} \n 分析完成，请查看！\n\n qcfail: \n {qcfail}\n')
+
     stdout.close()
     stderr.close()
 

wdl/postprocess.wdl

@@ -22,6 +22,7 @@ task run_post {
         fi
         indication.pl ${output_dir} ${cancer} ${project}
         sample_post.py -s ${name} -o ${output_dir}
+        qc_check.py -n ${output_dir}/qc/${name}_qc.txt -s ${output_dir}/qc/${normal}_qc.txt -p ${output_dir}/pollution/${name}_gender_res.txt -c ${output_dir}/qc/${name}_post.json -o ${output_dir}/report/${name}_qcfail.txt
         postprocess.py -n ${name} -s ${normal} -c ${output_dir} -o ${output_dir}/report/${name}.merged_file.xlsx
 
     }