多肽描述符计算：modlamp 包

_森罗万象

已于 2022-11-15 19:57:59 修改

阅读量907

点赞数

分类专栏：多肽/蛋白质处理文章标签：多肽描述符计算化学

于 2022-11-14 00:06:24 首次发布

本文链接：https://blog.csdn.net/weixin_52812620/article/details/127839893

版权

多肽/蛋白质处理专栏收录该内容

2 篇文章 0 订阅

订阅专栏

文章目录

1.GlobalDescriptor
2.PeptideDescriptor

本文主要讲 modlamp 的多肽描述符计算部分，文档在这，挑选 modlamp.descriptors 较难理解的部分

1.GlobalDescriptor

这个类根据序列计算每个多肽的全局描述符

from modlamp.descriptors import *
data=['AFDGHLKI','KKLQRSDLLRTK','KKLASCNNIPPR']
desc=GlobalDescriptor(data)

desc 对象存储了多肽序列库 (data) 的描述符矩阵，通过调用 desc 对象的方法来得到不同的描述符，存在 desc.descriptor 属性里

1.1.formula

desc.formula(amide=True,append=False) #amide表示C端是否酰胺化
desc.descriptor
"""
array([['C42 H66 N12 O10'],
       ['C64 H121 N23 O17'],
       ['C57 H102 N20 O15 S1']], dtype='<U19')
"""

计算分子式，amide 表示多肽序列的 C 端是否酰胺化，append 表示是否将此次得到的计算符添加到已有计算符矩阵中

desc.formula(amide=False,append=False) #可以看到N,H原子数少了,O原子数多了,因为 CONH2 -> COOH
desc.descriptor
"""
array([['C42 H65 N11 O11'],
       ['C64 H120 N22 O18'],
       ['C57 H101 N19 O16 S1']], dtype='<U19')
"""

将 amide 设为 False 之后，多肽序列的分子式改变了，因为从 CONH₂ 变成了 COOH，净增一个 O，减少一个N 和 H，可以从结果中看出

1.2.calculate_charge

这个描述符不太理解，参考的链接也打不开，下面是源码

def calculate_charge(self, ph=7.0, amide=False, append=False):
	desc = []
    for seq in self.sequences:
        desc.append(_charge(seq, ph, amide))  # calculate charge with helper function
    desc = np.asarray(desc).reshape(len(desc), 1)
    if append:
        self.descriptor = np.hstack((self.descriptor, np.array(desc)))
        self.featurenames.append('Charge')
    else:
        self.descriptor = np.array(desc)
        self.featurenames = ['Charge']
        
def _charge(seq, ph=7.0, amide=False):
    if amide:
        pos_pks = {'Nterm': 9.38, 'K': 10.67, 'R': 12.10, 'H': 6.04}
        neg_pks = {'Cterm': 15., 'D': 3.71, 'E': 4.15, 'C': 8.14, 'Y': 10.10}
    else:
        pos_pks = {'Nterm': 9.38, 'K': 10.67, 'R': 12.10, 'H': 6.04}
        neg_pks = {'Cterm': 2.15, 'D': 3.71, 'E': 4.15, 'C': 8.14, 'Y': 10.10}
    
    aa_content = count_aas(seq, scale='absolute')
    aa_content['Nterm'] = 1.0
    aa_content['Cterm'] = 1.0
    pos_charge = 0.0
    for aa, pK in pos_pks.items():
        c_r = 10 ** (pK - ph)
        partial_charge = c_r / (c_r + 1.0)
        pos_charge += aa_content[aa] * partial_charge
    neg_charge = 0.0
    for aa, pK in neg_pks.items():
        c_r = 10 ** (ph - pK)
        partial_charge = c_r / (c_r + 1.0)
        neg_charge += aa_content[aa] * partial_charge
    return round(pos_charge - neg_charge, 3)

其中 count_aas 效果如下，应该是计数各种氨基酸：

seq='AFDGHLKIA'
count_aas(seq, scale='absolute')
"""
OrderedDict([('A', 2.0),
             ('C', 0.0),
             ('D', 1.0),
             ('E', 0.0),
             ('F', 1.0),
             ('G', 1.0),
             ('H', 1.0),
             ('I', 1.0),
             ('K', 1.0),
             ('L', 1.0),
             ('M', 0.0),
             ('N', 0.0),
             ('P', 0.0),
             ('Q', 0.0),
             ('R', 0.0),
             ('S', 0.0),
             ('T', 0.0),
             ('V', 0.0),
             ('W', 0.0),
             ('Y', 0.0)])
"""

1.3.isoelectric_point

等电点

desc.isoelectric_point()
desc.descriptor
"""
array([[ 7.703125  ],
       [11.49511719],
       [10.68164062]])
"""

1.4.instability_index

不稳定指数，计算方法来自 Correlation between stability of a protein and its dipeptide composition: a novel approach for predicting in vivo stability of a protein from its primary sequence

desc.instability_index()
desc.descriptor
"""
array([[ 2.95      ],
       [81.925     ],
       [73.24166667]])
"""

1.5.aliphatic_index

脂肪族氨基酸指数

desc.aliphatic_index()
desc.descriptor
"""
array([[110.        ],
       [ 97.5       ],
       [ 73.33333333]])
"""

1.6.boman_index

蛋白相互作用的一种度量，计算方法来自 Antibacterial and antimalarial properties of peptides that are cecropin-melittin hybrids

desc.boman_index()
desc.descriptor
"""
array([[0.42      ],
       [4.33      ],
       [2.48083333]])
"""

1.7.hydrophobic_ratio

疏水性

desc.hydrophobic_ratio()
desc.descriptor
"""
array([[0.5       ],
       [0.25      ],
       [0.33333333]])
"""

1.8.calculate_all

计算实现的所有描述符

desc.calculate_all()
desc.descriptor,desc.featurenames
"""
(array([[8.00000000e+00, 8.99060000e+02, 1.03100000e+00, 1.14675328e-03,
         1.00253906e+01, 2.95000000e+00, 1.25000000e-01, 1.10000000e+02,
         4.20000000e-01, 5.00000000e-01],
        [1.20000000e+01, 1.48480000e+03, 4.98800000e+00, 3.35937500e-03,
         1.21787109e+01, 8.19250000e+01, 0.00000000e+00, 9.75000000e+01,
         4.33000000e+00, 2.50000000e-01],
        [1.20000000e+01, 1.33962000e+03, 3.83500000e+00, 2.86275212e-03,
         1.15722656e+01, 7.32416667e+01, 0.00000000e+00, 7.33333333e+01,
         2.48083333e+00, 3.33333333e-01]]),
 ['Length',
  'MW',
  'Charge',
  'ChargeDensity',
  'pI',
  'InstabilityInd',
  'Aromaticity',
  'AliphaticInd',
  'BomanInd',
  'HydrophRatio'])

"""

2.PeptideDescriptor

from modlamp.descriptors import *
data=['AFDGHLKI','KKLQRSDLLRTK','KKLASCNNIPPR']
desc=PeptideDescriptor(data,scalename='pepcats')
desc.descriptor,desc.sequences
"""
(array([], shape=(1, 0), dtype=float64),
 ['AFDGHLKI', 'KKLQRSDLLRTK', 'KKLASCNNIPPR'])
"""

scalename 可以选择不同的 amino acid descriptor scales，具体可见这里
在这里插入图片描述

2.1.calculate_crosscorr

desc.calculate_crosscorr(window=2,append=False),desc.descriptor,desc.descriptor.shape
"""
(None, array([[0.75      , 0.57142857, 0.25      , 0.14285714, 0.        ,
         0.14285714, 0.25      , 0.14285714, 0.25      , 0.14285714,
         0.        , 0.14285714, 0.25      , 0.        , 0.        ,
         0.14285714, 0.125     , 0.        , 0.125     , 0.        ,
         0.        , 0.14285714, 0.125     , 0.        , 0.        ,
         0.        , 0.        , 0.        , 0.125     , 0.        ,
         0.25      , 0.        , 0.25      , 0.        , 0.        ,
         0.        , 0.25      , 0.        , 0.        , 0.        ,
         0.125     , 0.        ],
        [0.66666667, 0.36363636, 0.        , 0.        , 0.        ,
         0.27272727, 0.41666667, 0.45454545, 0.41666667, 0.18181818,
         0.        , 0.        , 0.        , 0.        , 0.        ,
         0.        , 0.        , 0.        , 0.        , 0.        ,
         0.        , 0.        , 0.33333333, 0.09090909, 0.25      ,
         0.18181818, 0.        , 0.18181818, 0.08333333, 0.09090909,
         0.66666667, 0.45454545, 0.41666667, 0.27272727, 0.        ,
         0.09090909, 0.41666667, 0.09090909, 0.        , 0.        ,
         0.08333333, 0.        ],
        [0.75      , 0.54545455, 0.        , 0.        , 0.08333333,
         0.18181818, 0.33333333, 0.36363636, 0.25      , 0.18181818,
         0.        , 0.        , 0.        , 0.        , 0.        ,
         0.        , 0.        , 0.        , 0.        , 0.        ,
         0.        , 0.        , 0.33333333, 0.27272727, 0.33333333,
         0.27272727, 0.        , 0.        , 0.        , 0.        ,
         0.58333333, 0.36363636, 0.25      , 0.09090909, 0.        ,
         0.        , 0.25      , 0.09090909, 0.        , 0.        ,
         0.        , 0.        ]]), (3, 42))
"""

2.2.calculate_autocorr

desc.calculate_autocorr(window=2,append=False),desc.descriptor,desc.descriptor.shape
"""
(None, array([[0.75      , 0.25      , 0.125     , 0.25      , 0.25      ,
         0.125     , 0.57142857, 0.        , 0.        , 0.        ,
         0.        , 0.        ],
        [0.66666667, 0.        , 0.33333333, 0.66666667, 0.41666667,
         0.08333333, 0.36363636, 0.        , 0.09090909, 0.45454545,
         0.09090909, 0.        ],
        [0.75      , 0.        , 0.33333333, 0.58333333, 0.25      ,
         0.        , 0.54545455, 0.        , 0.27272727, 0.36363636,
         0.09090909, 0.        ]]), (3, 12))
"""

2.3.calculate_moment

data=['AFDGHLKI','KKLQRSDLLRTK','KKLASCNNIPPR']
desc=PeptideDescriptor(data,'eisenberg')
desc.descriptor,desc.sequences
desc.calculate_moment()
desc.descriptor
"""
array([[0.4205886 ],
       [0.20845564],
       [0.41395658]])
"""

计算这个描述符的时候选择某些 amino acid descriptor scales 会报错

2.4.calculate_global

data=['AFDGHLKI','KKLQRSDLLRTK','KKLASCNNIPPR']
desc=PeptideDescriptor(data,'pepcats')
desc.descriptor,desc.sequences
desc.calculate_global()
desc.descriptor
"""
array([[1.75      ],
       [2.16666667],
       [1.91666667]])
"""