从蛋白质结构文件pdb中提取序列和骨架原子坐标

pdb文件是蛋白质结构存储的标准文件，用txt打开其主要内容如下所示

上图部分是model1下的蛋白原子分布，序号1所在列是原子序号，序号2所在列是氨基酸种类，序号3所在列是氨基酸序号，序号4所在三列是原子的三维坐标，序号5所在列是原子种类

用python实现提取序列：

from Bio import PDB
本文件用于提取单个pdb文件的氨基酸
parser = PDB.PDBParser()
io = PDB.PDBIO()
fileName_save = HHH_b1_00224.fasta

nameOfFile_read = HHH_b1_00224.pdb
下面将氨基酸全称改为简写
longToShort = {GLY: G, ALA: A, VAL: V, LEU: L, ILE: I, PHE: F, TRP: W, TYR: Y,ASP: D ,HIS: H, ASN: N, GLU: E, LYS: K, GLN: Q, MET: M, ARG: R, SER: S, THR: T, CYS: C, PRO: P, SEC: U, PYL: O}

struct = parser.get_structure(aaaa, nameOfFile_read) 此处的aaaa是结构名，无实际意义
    with open(fileName_save, a+) as f_w:
        f_w.write(>sequence + str(i+1) + 
)
        for model in struct: #遍历所有model
            for chain in model: #遍历model中的所有chain
                for residue in chain: #遍历chain中的所有残基
                    resname = residue.get_resname() #读取氨基酸
                    f_w.write(longToShort[resname])

若要读取原子坐标，只需在 for residue in chain下增加一个遍历所有原子即可，下述代码实现读取骨架原子坐标

for model in struct:
    for chain in model:
        for residue in chain:
            for atom in residue:
                if(num < 3):
                    x,y,z = atom.get_coord()
                    with open(filename, a+) as file_object:
                        #file_object.write(NaN + ", " + NaN + ", " + NaN + "], [")
                        file_object.write(str(x) + ", " + str(y) + ", " + str(z) + "], [")
                    num = num + 1
                elif(num == 3):
                    x, y, z = atom.get_coord()
                    with open(filename, a+) as file_object:
                        #file_object.write(NaN + ", " + NaN + ", " + NaN + "]]")
                        file_object.write(str(x) + ", " + str(y) + ", " + str(z) + "]]")
                    num = num + 1