#!/usr/bin/env python import sys import numpy as np import scipy.spatial.distance as spd import scipy.cluster.hierarchy as sph from scipy import stats import matplotlib #matplotlib.use('Agg') import pylab import pandas as pd import matplotlib.pyplot as plt class ReadCmd: def __init__( self ): import argparse as ap import textwrap p = ap.ArgumentParser( description= "TBA" ) arg = p.add_argument arg( '-i', '--inp', '--in', metavar='INPUT_FILE', type=str, nargs='?', default=sys.stdin, help= "The input matrix" ) arg( '-o', '--out', metavar='OUTPUT_FILE', type=str, nargs='?', default=None, help= "The output image file [image on screen of not specified]" ) arg( '-m', '--metadata_file', type=str, default='None', help= "The input metadata file [default None]" ) DataMatrix.input_parameters( p ) BarPlot.input_parameters( p ) self.args = p.parse_args() def check_consistency( self ): pass def get_args( self ): return self.args class DataMatrix: datatype = 'data_matrix' @staticmethod def input_parameters( parser ): dm_param = parser.add_argument_group('Input data matrix parameters') arg = dm_param.add_argument arg( '--sep', type=str, default='\t' ) arg( '-f', '--feat', type=str, default=None, required = True, help = "Name of the feature to plot" "[or the ending string if --endswith is specified]") arg( '--endswith', action='store_true', help = "Match the ending part of the feature name" ) arg( '--fname_row', type=int, default=0, help = "row number containing the names of the features " "[default 0, specify -1 if no names are present in the matrix") arg( '--sname_row', type=int, default=0, help = "column number containing the names of the samples " "[default 0, specify -1 if no names are present in the matrix") arg( '--skip_rows', type=str, default=None, help = "Row numbers to skip (0-indexed, comma separated) from the input file" "[default None, meaning no rows skipped") arg( '--def_na', type=float, default=None, help = "Set the default value for missing values [default None which means no replacement]") def __init__( self, input_file, args ): self.args = args toskip = [int(l) for l in self.args.skip_rows.split(",")] if self.args.skip_rows else None self.table = pd.read_table( input_file, sep = self.args.sep, skipinitialspace = True, skiprows = toskip, header = self.args.fname_row if self.args.fname_row > -1 else None, index_col = self.args.sname_row if self.args.sname_row > -1 else None ) rows = [] if self.args.endswith: for n in self.table.index: if n.endswith( self.args.feat ): rows.append( n ) elif self.args.feat in self.table.index: rows.append( self.args.feat ) self.table = self.table.reindex( index=rows ) if not len(rows): sys.stderr.write("Error, feat "+self.args.feat+" not found!") sys.exit() if len(rows) > 1: sys.stderr.write("Error, multiple features matching "+self.args.feat+" !") sys.exit() if not self.args.def_na is None: self.table = self.table.fillna( self.args.def_na ) def get_numpy_matrix( self ): return self.table def get_snames( self ): return list(self.table.index) def get_fnames( self ): return list(self.table.columns) def save_matrix( self, output_file ): self.table.to_csv( output_file, sep = '\t' ) class MetadataMatrix: datatype = 'metadata_matrix' @staticmethod def input_parameters( parser ): dm_param = parser.add_argument_group('Input metadata file') arg = dm_param.add_argument arg( '--sep', type=str, default='\t' ) arg( '--fname_row', type=int, default=0, help = "row number containing the names of the features " "[default 0, specify -1 if no names are present in the matrix") arg( '--def_na', type=float, default=None, help = "Set the default value for missing values [default None which means no replacement]") def __init__( self, input_file, args ): self.args = args self.table = pd.read_table( input_file, sep = self.args.sep, skipinitialspace = True, #header = self.args.fname_row if self.args.fname_row > -1 else None, index_col = self.args.sname_row if self.args.sname_row > -1 else None ) if not self.args.def_na is None: self.table = self.table.fillna( self.args.def_na ) def get_snames( self ): return list(self.table.index) def get_fnames( self ): return list(self.table.columns) def get_table( self ): return self.table class BarPlot: datatype = 'barplot' @staticmethod def input_parameters( parser ): hm_param = parser.add_argument_group('Heatmap options') arg = hm_param.add_argument arg( '--dpi', type=int, default=72, help = "Image resolution in dpi [default 72]") arg( '-C', '--color_condition', type=str, default=None, help = "The name of the metadata column used for coloring") arg( '-H', '--hatch_condition', type=str, default=None, help = "The name of the metadata column used for hatching") arg( '-G', '--group_condition', type=str, default=None, help = "The name of the metadata column used for grouping") arg( '-t', '--title', type=str, default=None, help = "The title of the plot [default no title]") arg( '-l', '--log_scale', action='store_true', help = "Log scale" ) def __init__( self, numpy_matrix, metadata_matrix, args = None ): self.numpy_matrix = numpy_matrix self.mmatrix = metadata_matrix self.args = args def draw( self ): fig = plt.figure( figsize=(20,8) ) ax = fig.add_subplot(111) width = 0.65 names = list(self.numpy_matrix.index) n0 = names[0] tp = self.numpy_matrix.to_dict() keys = sorted(tp) if self.args.color_condition not in self.mmatrix: self.args.color_condition = None cond_values = [None] if self.args.color_condition is None else sorted(set(self.mmatrix[self.args.color_condition]) ) if self.args.hatch_condition not in self.mmatrix: self.args.hatch_condition = None hatch_values = [None] if self.args.hatch_condition is None else sorted(set(self.mmatrix[self.args.hatch_condition]) ) if self.args.group_condition: group_values = list(sorted(set(self.mmatrix[self.args.group_condition]))) keys = sorted( keys, key=lambda x:group_values.index(self.mmatrix[self.args.group_condition][x]) ) else: keys, group_values = sorted( keys ), [] ind = np.arange( len(tp) ) pos = ind-width/2 hatches = ['//','\\\\','++','--','xx'] cols = ['r','g','c','b'] minv,maxv = 0.0, max([v[n0] for v in tp.values()]) bar_sets = [] for i,c in (enumerate(cond_values) if len(cond_values) > 0 else None): for j,h in enumerate(hatch_values): values = [(tp[k][n0] if (c is None or self.mmatrix[self.args.color_condition][k] == c) and (h is None or self.mmatrix[self.args.hatch_condition][k] == h) else 0.0) for k in keys] b = ax.bar(pos, values, width, hatch=hatches[j%len(hatches)] if len(hatch_values) > 1 else "", color=cols[i%len(cols)]) cond = self.args.color_condition + " "+str(c).strip()+", " if c else "" hatch = self.args.hatch_condition + " "+str(h).strip()+", " if h else "" bar_sets.append( (b,cond+hatch) ) v0 = ind[0]-0.5 vm1 = v0 ax.plot([v0,v0],[minv,maxv],"--",linewidth=2,color='k') for g in group_values: vm1 = v0 v0 += list(self.mmatrix[self.args.group_condition]).count(g) ax.plot([v0,v0],[minv,maxv],"--",linewidth=2,color='k') ax.text( (vm1+v0)*0.5, maxv * 0.9, str(g), horizontalalignment='center', verticalalignment='center' ) #ax.text( (vm1+v0)*0.5, maxv * 0.9, str(round(g,1)), horizontalalignment='center', verticalalignment='center' ) if self.args.color_condition or self.args.hatch_condition: leg = ax.legend( zip(*bar_sets)[0], zip(*bar_sets)[1], bbox_to_anchor=(1.02, 0,0.3,1), loc=1, ncol=1, mode="expand", borderaxespad=0., frameon = False) ax.set_xlim(-width,ind[-1]+width) ax.set_ylim(0,maxv) ax.set_xticks( ind ) ax.set_xticklabels( keys, rotation = 90 ) ax.set_title( self.args.title or "" ) if not self.args.out: plt.show() else: fig.savefig( self.args.out, bbox_inches='tight', dpi = self.args.dpi, bbox_extra_artists=((fig.get_axes()[0].get_legend(),) if self.args.color_condition or self.args.hatch_condition else None) ) #dpi = self.args.dpi ) if __name__ == '__main__': read = ReadCmd( ) read.check_consistency() args = read.get_args() dm = DataMatrix( args.inp, args ) mdm = MetadataMatrix( args.metadata_file, args ) bp = BarPlot( dm.get_numpy_matrix(), mdm.get_table(),args ) bp.draw()