update last

diff.ipynb

@@ -591,6 +591,37 @@
    "metadata": {},
    "outputs": [],
    "source": []
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "2024-03-07 \n",
+    "['5ksa', '1g6r', '6bga', '6u3n', '1zgl', '7l1d', '3qiu', '2ypl', '7z50', '6uln', '1mwa', '4z7v', '4ozi', '7rrg', '6uk4', '4p2o', '6vm8', '5ksb'] 重新建模\n",
+    "1g6r: Align 对齐过程中，序列编号出错导致残基缺失，进而导致MD模拟后续失败\n",
+    "6u3n: A开头地方与B链的碰撞\n",
+    "6uk4: 结构分散（不适合模拟）\n",
+    "6uln: 结构分散（不适合模拟）\n",
+    "6vm8: 结构分散（不适合模拟）\n",
+    "7l1d: 无碰撞，结构分散（不适合模拟）\n",
+    "7rrg: 无碰撞，结构分散（不适合模拟）\n",
+    "7z50: A与B链的碰撞\n",
+    "1mwa: B链116缺失\n",
+    "1zgl: M链122,123缺失\n",
+    "2ypl: D链124E链121碰撞\n",
+    "3qiu: A与B链的碰撞\n",
+    "4ozi: A链72位与B链5位（β折叠）碰撞\n",
+    "4p2o: A链67位与B链4位碰撞\n",
+    "4z7v: C链96位与D链105碰撞\n",
+    "5ksa: 缺失过多导致修复后loop过多（不适合模拟）\n",
+    "5ksb: C链83位与D链5位碰撞\n",
+    "6bga: B链137，138虚线缺失\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
   }
  ],
  "metadata": {

fixsequence.py

@@ -0,0 +1,23 @@
+#!/usr/bin/env python
+# -*- encoding: utf-8 -*-
+'''
+@file               :fixsequence.py
+@Description:       : fix sequence of pdb file
+@Date               :2024/03/07 09:44:06
+@Author             :lyzeng
+@Email              :pylyzeng@gmail.com
+@version            :1.0
+'''
+from analysis_pdb import PDBAnalyzer
+from pathlib import Path
+# Specify the path to your PDB file
+pdb_files = [
+    Path('./pdb_test9/1zgl.pdb'),
+    Path('./pdb_test9/1g6r.pdb'),
+    Path('./pdb_test9/1mwa.pdb'),
+    Path('./pdb_test9/6bga.pdb'),
+]
+for pdb_file in pdb_files:
+    # 修复所有链的编号
+    pdb_analyzer_instance = PDBAnalyzer.renumber_residues_based_on_issues_and_clean(pdb_file)
+    print(pdb_file.stem, pdb_analyzer_instance.renumber_errors)

manualfix/README.md

@@ -0,0 +1,66 @@
+2024-03-07 
+['5ksa', '1g6r', '6bga', '6u3n', '1zgl', '7l1d', '3qiu', '2ypl', '7z50', '6uln', '1mwa', '4z7v', '4ozi', '7rrg', '6uk4', '4p2o', '6vm8', '5ksb'] 重新建模
+# 1g6r: Align 对齐过程中，序列编号出错导致残基缺失，进而导致MD模拟后续失败
+# 6u3n: A开头地方与B链的碰撞
+6uk4: 结构分散（不适合模拟）
+6uln: 结构分散（不适合模拟）
+6vm8: 结构分散（不适合模拟）
+7l1d: 无碰撞，结构分散（不适合模拟）
+7rrg: 无碰撞，结构分散（不适合模拟）
+# 7z50: A与B链的碰撞
+# 1mwa: B链116缺失
+# 1zgl: M链122,123缺失
+# 2ypl: D链124E链121碰撞
+# 3qiu: A与B链的碰撞
+# 4ozi: A链72位与B链5位（β折叠）碰撞
+# 4p2o: A链67位与B链4位碰撞
+# 4z7v: C链96位与D链105碰撞
+5ksa: 缺失过多导致修复后loop过多（不适合模拟）
+# 5ksb: C链83位与D链5位碰撞
+# 6bga: B链137，138虚线缺失
+
+最终确定：['6bga', '5ksb', '4z7v', '4p2o', '4ozi', '3qiu', '2ypl', '1zgl', '1mwa', '7z50', '6u3n', '1g6r'] 
+
+____
+1zgl [{'chain_id': 'P', 'start_residue': 63, 'end_residue': 104, 'estimated_missing': 0}] B链M链缺失太多，修复之后loop过多，不适合模拟
+1g6r [{'chain_id': 'B', 'start_residue': 109, 'end_residue': 237, 'estimated_missing': 0}] 修复
+1mwa [{'chain_id': 'B', 'start_residue': 116, 'end_residue': 316, 'estimated_missing': 0}] 多处编号错误B链
+6bga [{'chain_id': 'B', 'start_residue': 112, 'end_residue': 198, 'estimated_missing': 0}] B链两处缺失虚线
+____
+放手动修复的pdb文件，再提取单聚体后手动修复残缺编号。
+
+1g6r.manualfix.pdb 
+1mwa.manualfix.pdb
+6bga.modellerfix.pdb B链碰撞导致缺失
+cp ../pdb_test7/runner_5ksb/5ksb.modellerfix.pdb ./
+
+## pyrosetta fastrelax
+
+```
+# 使用pyrosetta快速修复蛋白缺失的侧链
+from pathlib import Path
+import pyrosetta
+from pyrosetta import rosetta
+from multiprocessing import Pool
+
+
+def fix_optimize(file: Path, out_file: Path):
+    '''
+    FastRelax使用快速梯度下降算法，可以在较短时间内对蛋白质进行优化，并且对于结构中的非构象缺陷，例如氢键、离子对、芳香性相互作用和溶剂-蛋白质相互作用等进行优化。
+    ref2015是Rosetta程序包中的一个分数函数，它是Rosetta2015中引入的一个新的蛋白质力场，用于蛋白质结构预测和设计。这个力场是从先前的Rosetta力场中提炼出来的，经过了一系列的校正和优化，可以更好地预测蛋白质的折叠构象。在FastRelax中，ref2015可以作为一个可选参数来指定使用哪个力场来进行优化。
+    :param file:
+    :param out_file:
+    :return: 
+    '''
+    # 使用pyrosetta修复蛋白结构
+    # 初始化PyRosetta
+    pyrosetta.init()
+    # 读入蛋白质结构
+    pose = pyrosetta.pose_from_pdb(file.as_posix())
+    # fix residue side chain
+    scorefxn = pyrosetta.create_score_function('ref2015')
+    relax = pyrosetta.rosetta.protocols.relax.FastRelax(scorefxn)
+    relax.apply(pose)
+    # 输出修复后的结构
+    pose.dump_pdb(out_file.as_posix())
+```

manualfix/relax.py

@@ -0,0 +1,46 @@
+from pathlib import Path
+import pyrosetta
+from multiprocessing import Pool
+from shutil import copyfile
+import logging
+
+# fix = ['6bga', '5ksb', '4z7v', '4p2o', '4ozi', '3qiu', '2ypl', '1zgl', '7z50', '6u3n']
+fix = ['1zgl']
+
+def fix_optimize(file: Path, out_file: Path):
+    # 设置日志
+    log_file = out_file.with_suffix('.log')
+    logger = logging.getLogger(log_file.name)
+    logger.setLevel(logging.INFO)
+    formatter = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s')
+    file_handler = logging.FileHandler(log_file)
+    file_handler.setFormatter(formatter)
+    logger.addHandler(file_handler)
+
+    logger.info(f'Processing {file.name}')
+
+    pyrosetta.init()
+    pose = pyrosetta.pose_from_pdb(file.as_posix())
+    scorefxn = pyrosetta.create_score_function('ref2015')
+    relax = pyrosetta.rosetta.protocols.relax.FastRelax(scorefxn)
+    relax.apply(pose)
+    pose.dump_pdb(out_file.as_posix())
+
+    logger.info(f'Finished processing {file.name}')
+
+if __name__ == '__main__':
+    dir1 = Path('../pdb_test8')
+    dir2 = Path('../pdb_test7')
+    dirs = [dir1, dir2]
+    files = []
+    for dir in dirs:
+        files.extend(list(dir.rglob('*.modellerfix.pdb')))
+    for file in files:
+        if file.name.split('.')[0] in fix:
+            print(file)
+            target = Path('/mnt/mydrive/analysis_pdb-dev/manualfix') / file.name
+            copyfile(file.as_posix(), target.as_posix())
+    pyrosetta_fix = list(Path('/mnt/mydrive/analysis_pdb-dev/manualfix').rglob('*.modellerfix.pdb'))
+    with Pool(16) as p:
+        p.starmap(fix_optimize, [(file, file.with_stem(file.stem + '.fastrelax')) for file in pyrosetta_fix])
+    print('fastrelax done')