NeuroMario/Evolution.py at master · Ashafix/NeuroMario · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
class Evolution:
    def __init__(self):
        self.solution = list()
        self.generations = dict()

    def mate_solutions(self, sol1, sol2, prob=0.5):
        keys = list(sol1.keys())
        keys.extend(list(sol2.keys()))
        child_solution = dict()
        for key in keys:
            if key in sol1 and key in sol2:
                if random.random() < prob:
                    child_solution[key] = sol1[key]
                else:
                    child_solution[key] = sol2[key]
            elif key in sol1:
                child_solution[key] = sol1[key]
            else:
                child_solution[key] = sol2[key]
        return child_solution

    def trim_genome(self, genome):
        output_genome = dict()
        for gene in genome:
            output_genome[gene] = genome[gene][0]
        return output_genome

    def mutate_solution(self, solution, prob=0.1, allowed=b'|..|UDLRs.YBXAlr|............|'):
        mutated_solution = dict()
        for key in solution:

            if random.random() < prob:
                print()
                print('mutate', key)
                pos = random.randint(4, 15)
                while pos in (9,):
                    pos = random.randint(4, 15)
                print(solution[key][pos:pos + 1], allowed[pos:pos + 1], pos)
                print('before', solution[key])

                if solution[key][pos:pos + 1] == allowed[pos:pos + 1]:
                    print('silencing')
                    mutated_solution[key] = solution[key][0:pos] + b'.' + solution[key][pos + 1:]
                else:
                    print('gain of function')
                    mutated_solution[key] = solution[key][0:pos] + allowed[pos:pos + 1] + solution[key][pos + 1:]
                print('after ', mutated_solution[key])
            else:
                mutated_solution[key] = solution[key]
        return mutated_solution

    def cross_solutions(self, sol1, sol2, mutate=0.1):
        new_sol = self.mate_solutions(sol1, sol2)
        if mutate > 0:
            new_sol = self.mutate_solution(new_sol, prob=mutate)

        return new_sol


class Organism:
    def __init__(self, name, sequence, run_times=list()):
        self.name = name
        self.sequence = sequence
        self.run_times = run_times


class Generation:
    def __init__(self):
        self.parents = dict()
        self.children = dict()
        self.performance = dict()
        self.statistics = dict()
        self.generation_number = 0
        self.name = ''
        self.similarity = dict()

    def add_parent(self, parent):
        if parent.name in self.parents:
            return False
        self.parents[parent.name] = parent
        return True

    def add_child(self, parent):
        if children.name in self.children:
            return False
        self.children[children.name] = children
        return True

    def get_child_statistics(self):
        if not self.statistics.get('children'):
            self.calculate_statistics()
        return self.statistics['children']

    def get_parent_statistics(self):
        if not self.statistics.get('parents') or not self.statistics.get('children'):
            self.calculate_statistics()
        return self.statistics['parents']

    def calculate_statistics(self):

        self.statistics['children'] = dict()
        self.statistics['parents'] = dict()

        for child_name, child in self.children.items():
            self.statistics['children'][child_name] = dict()
            self.statistics['children'][child_name]['run_times'] = child.run_times
            self.statistics['children'][child_name]['avg_run_times'] = statistics.mean(child.run_times)
            self.statistics['children'][child_name]['med_run_times'] = statistics.median(child.run_times)

        for parent_name, parent in self.parents.items():
            self.statistics['parents'][parent_name] = dict()
            self.statistics['parents'][parent_name]['run_times'] = parent.run_times
            self.statistics['parents'][parent_name]['avg_run_times'] = statistics.mean(parent.run_times)
            self.statistics['parents'][parent_name]['med_run_times'] = statistics.median(parent.run_times)

        self.statistics['summary']['avg_run_times_children'] = statistics.mean(
            self.statistics['children'][children_name]['avg_run_times'] for children_name in self.children)
        self.statistics['summary']['avg_run_times_parents'] = statistics.mean(
            self.statistics['parents'][parent_name]['avg_run_times'] for parent_name in self.parents)
        self.statistics['summary']['med_run_times_children'] = statistics.median(
            self.statistics['children'][children_name]['avg_run_times'] for children_name in self.children)
        self.statistics['summary']['med_run_times_parents'] = statistics.median(
            self.statistics['parents'][parent_name]['avg_run_times'] for parent_name in self.parents)

        self.statistics['comparison'][child_name]['avg_run_times'] = self.statistics['summary'][
                                                                         'avg_run_times_children'] / \
                                                                     self.statistics['summary']['avg_run_times_parents']
        self.statistics['comparison'][child_name]['med_run_times'] = self.statistics['summary'][
                                                                         'med_run_times_children'] / \
                                                                     self.statistics['summary']['med_run_times_parents']

    def calculate_similarities(self):

        for parent_name, parent in self.parents.items():
            for parent2_name, parent2 in self.parents.items():
                if parent_name == parent2_name:
                    self.similarity[parent_name] = dict()
                    self.similarity[parent2_name] = dict()
                    self.similarity[parent_name][parent2_name] = dict(homology=0, new_genes=0)
                    self.similarity[parent2_name][parent_name] = dict(homology=0, new_genes=0)
                self.calculate_similarity(parent, parent2)

    def calculate_diff(self, sequence1, sequence2):
        diff = 0
        if len(sequence1) != len(sequence2):
            return 1
        for i, s in enumerate(sequence1):
            if s != sequence2[i]:
                diff += 1
        diff /= len(sequence1)
        return diff

    def calculate_similarity(self, gene1, gene2):

        homology = 0
        new_genes = 0
        for gene in gene1.sequence:
            if gene in gene2.sequence:
                homology += self.calculate_diff(gene1.sequence[gene], gene2.sequence[gene])
            else:
                new_genes += 1
        for gene in gene2.sequence:
            if gene not in gene1.sequence:
                new_genes += 1

        if not self.similarity.get(gene1.name):
            self.similarity[gene1.name] = dict()

        self.similarity[gene1.name][gene2.name] = dict(homology=homology, new_genes=new_genes)

        if not self.similarity.get(gene2.name):
            self.similarity[gene2.name] = dict()

        self.similarity[gene2.name][gene1.name] = dict(homology=homology, new_genes=new_genes)