Source code for hylite.SNP

#!/usr/bin/env python3

#    (c) Copyright 2013-2018 Murray Cox, Wandrille Duchemin, Pierre-Yves Dupont.
#
#
#    This file is part of HyLiTE.
#
#    HyLiTE is a free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License version 3 as published by
#    the Free Software Foundation.
#
#    HyLiTE is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with HyLiTE.  If not, see <http://www.gnu.org/licenses/>


#===============================#
# author: Wandrille Duchemin    #
#         Murray Cox            #
# last modified: 9 January 2018 #
#===============================#


class SNP:
    '''
    This class is used to model a SNP, including its position on the reference genome and
    the presence/absence/coverage of a SNP in multiple organisms
    
    Attributes:
         - gene (str): the gene containing the snp
         - position (int): the position of the snp on the gene
         - ref (str): the reference allele
         - alt (str): the alternative allele
         - masked (bool): a boolean indicating if at least one organism has a bad coverage at the snp position
         - presence (dict): key is organism name, value a tuple containing is 1 if SNP is present on a gene copy, 0 if absent, -1 if the coverage is bad
    '''
    def __init__(self, gene, position, ref, alt):
        '''
        Args:
             - gene (str): name of a gene
             - position (int): position on the gene
             - ref (str): the reference allele
             - alt (str): the alternative allele
        '''
        self.gene=gene
        self.position=position
        self.ref=ref
        self.alt=alt
        self.masked = False #is True if at least one of the organism has bad coverage at <position> of <gene>
        self.presence = {} #key is organism name, value is tuple containing 1 if SNP is present on a gene copy, 0 if absent, -1 if the coverage is bad
        return

    def has_SNP(self, orgname, i):
        '''Update the SNP with its presence in an organism
        
        Args:
             - orgname (str): name of an organism
             - i (int): gene copy the snp is present on, OR -1 if it is the child
        '''
        if i == -1:#this is the child
            self.presence[orgname] = [1]

        else:#not the child
            if (orgname in self.presence) is False:#new organism
                self.presence[orgname] = list()#initialise

            while len(self.presence[orgname])<= i:#extend the list to desired size
                self.presence[orgname].append(0)#by default, we push 0s in the presence

            self.presence[orgname][i]=1#updating for presence
            
        return
    
    def no_SNP(self, orgname, i):
        '''Update the SNP with its absence in an organism
        
        Args:
             - orgname (str): name of an organism
             - i (int): gene copy the snp is present on, OR -1 if it is the child
        '''
        if i == -1:#this is the child
            if (orgname in self.presence) is False:#only if no presence or absence has been recorded yet
                self.presence[orgname] = [0]
            else:
                self.presence[orgname].append(0)

        else:#not the child
            if (orgname in self.presence) is False:#new organism
                self.presence[orgname] = list()#initialise

            while len(self.presence[orgname])<= i:#extend the list to desired size
                self.presence[orgname].append(0)#by default, we push 0s in the presence

            self.presence[orgname][i]=0#updating for absence
            
        return
        
    def no_coverage(self, orgname):
        '''Update the SNP with its bad coverage in an organism
        
        Args:
             - orgname (str): name of an organism
        '''
        self.presence[orgname] = [-1]
        self.masked = True
        return

    def listed_org(self):
        '''
        Returns:
             - list. the names of the organisms listed in self.presence
        '''
        return list(self.presence.keys())
    
    def __str__(self):
        return  '\t'.join([self.gene, str(self.position), self.alt, self.presence.__str__()])
        
    def __repr__(self):
        return self.__str__()