# HG changeset patch # User glogobyte # Date 1634141108 0 # Node ID d58f050acd1801b61eda761973f36bb9cb26ca92 # Parent 810e789ffeab6d633bd647f5fb5f44ba2a8f5261 Uploaded diff -r 810e789ffeab -r d58f050acd18 mirgene_graphs.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/mirgene_graphs.py Wed Oct 13 16:05:08 2021 +0000 @@ -0,0 +1,686 @@ +import itertools +import pandas as pd +from math import pi +import numpy as np +import matplotlib.pyplot as plt +import math +import logomaker as lm +from fpdf import FPDF, fpdf +import glob + +############################################################################################################################################################################### +def pie_non_temp(merge_con,merge_non_con,merge_tre,merge_non_tre,c_unmap,t_unmap,c_unmap_counts,t_unmap_counts,group_name1,group_name2): + + c_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_con] + t_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_tre] + c_non_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_non_con] + t_non_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_non_tre] + + c_templ = 0 + c_tem_counts = 0 + c_mature = 0 + c_mat_counts = 0 + t_templ = 0 + t_tem_counts = 0 + t_mature = 0 + t_mat_counts = 0 + + c_non = len(c_non_samples) + c_non_counts = sum(x[2] for x in c_non_samples) + t_non = len(t_non_samples) + t_non_counts = sum(x[2] for x in t_non_samples) + + c_unmap = c_unmap - c_non + t_unmap = c_unmap - t_non + + c_unmap_counts=c_unmap_counts - c_non_counts + t_unmap_counts=t_unmap_counts - t_non_counts + + + for x in c_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + else: + c_templ+=1 + c_tem_counts += x[2] + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + f=1 + break + if f==0: + c_templ+=1 + c_tem_counts += x[2] + + for x in t_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + else: + t_templ+=1 + t_tem_counts += x[2] + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + f=1 + break + if f==0: + t_templ+=1 + t_tem_counts += x[2] + + fig = plt.figure(figsize=(7,5)) + + labels = 'miRNA RefSeq','Template', 'Unassigned' ,'Non-template' + sizes = [c_mat_counts, c_tem_counts, c_unmap_counts,c_non_counts] + colors = ['gold', 'yellowgreen', 'lightcoral', 'lightskyblue'] + # Plot + ax1 = plt.subplot2grid((1,2),(0,0)) + patches, texts, autotexts=plt.pie(sizes, labels=labels, colors=colors, startangle=140,autopct='%1.1f%%',radius=0.8) + [x.set_fontsize(8) for x in texts] + plt.title(group_name1 + ' Group (reads)',fontsize=12) + + labels = 'miRNA RefSeq','Template', 'Unassigned','non-template' + sizes = [t_mat_counts, t_tem_counts, t_unmap_counts, t_non_counts] + colors = ['gold', 'yellowgreen', 'lightcoral', 'lightskyblue'] + # Plot + ax2 = plt.subplot2grid((1,2),(0,1)) + patches, texts, autotexts=plt.pie(sizes, labels=labels, colors=colors, startangle=140,autopct='%1.1f%%',radius=0.8) + [x.set_fontsize(8) for x in texts] + plt.title(group_name2 + ' Group (reads)', fontsize=12) + plt.savefig('pie_non.png',dpi=300) + +#################################################################################################################################################################################################################### + +def pie_temp(merge_con,c_unmap,c_unmap_counts,merge_tre,t_unmap,t_unmap_counts,group_name1,group_name2): + + c_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_con] + t_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_tre] + + c_templ = 0 + c_tem_counts = 0 + c_mature = 0 + c_mat_counts = 0 + t_templ = 0 + t_tem_counts = 0 + t_mature = 0 + t_mat_counts = 0 + + for x in c_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + else: + c_templ+=1 + c_tem_counts += x[2] + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + f=1 + break + if f==0: + c_templ+=1 + c_tem_counts += x[2] + + for x in t_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + else: + t_templ+=1 + t_tem_counts += x[2] + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + f=1 + break + if f==0: + t_templ+=1 + t_tem_counts += x[2] + + + fig = plt.figure() + labels = 'miRNA RefSeq','Template', 'Unassigned' + sizes = [c_mat_counts, c_tem_counts, c_unmap_counts] + colors = ['gold', 'yellowgreen', 'lightskyblue'] + explode = (0.2, 0.05, 0.1) + # Plot + ax1 = plt.subplot2grid((1,2),(0,0)) + patches, texts, autotexts=plt.pie(sizes, labels=labels, colors=colors, startangle=140,autopct='%1.1f%%',radius=0.8) + [x.set_fontsize(8) for x in texts] + plt.title(group_name1 + ' group (reads)', fontsize=12) + labels = 'miRNA RefSeq','Template', 'Unassigned' + sizes = [t_mat_counts, t_tem_counts, t_unmap_counts] + colors = ['gold', 'yellowgreen', 'lightskyblue'] + explode = (0.2, 0.05, 0.1) + # Plot + ax2 = plt.subplot2grid((1,2),(0,1)) + patches, texts, autotexts=plt.pie(sizes, labels=labels, colors=colors, startangle=140,autopct='%1.1f%%',radius=0.8) + [x.set_fontsize(8) for x in texts] + plt.title(group_name2 + ' group (reads)',fontsize = 12) + plt.savefig('pie_tem.png',dpi=300) + +################################################################################################################################################################################################################### + +def make_spider(merge_con,merge_tre,group_name1,group_name2): + + c_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_con] + t_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge_tre] + + c_5 = 0 + c_5_counts = 0 + c_3 = 0 + c_3_counts = 0 + c_both =0 + c_both_counts=0 + c_mature = 0 + c_mat_counts = 0 + c_exception=0 + c_exception_counts=0 + + + t_5 = 0 + t_5_counts = 0 + t_3 = 0 + t_3_counts = 0 + t_both = 0 + t_both_counts = 0 + t_mature = 0 + t_mat_counts = 0 + t_exception = 0 + t_exception_counts=0 + + for x in c_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + elif 0 == int(x[0].split("_")[3]): + c_5+=1 + c_5_counts += x[2] + elif 0 == int(x[0].split("_")[4]): + c_3+=1 + c_3_counts += x[2] + else: + c_both+=1 + c_both_counts+=x[2] + + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + c_mature+=1 + c_mat_counts += x[2] + f=1 + break + if f==0: + part=x[0].split("/")[0].split("_")[-2]+"_"+x[0].split("/")[0].split("_")[-1] + num=0 + for y in x[0].split("/"): + if y.split("_")[-2]+"_"+y.split("_")[-1]==part: + num+=1 + if num==len(x[0].split("/")): + + if 0 == int(x[0].split("/")[0].split("_")[3]): + c_5+=1 + c_5_counts += x[2] + elif 0 == int(x[0].split("/")[0].split("_")[4]): + c_3+=1 + c_3_counts += x[2] + else: + c_both+=1 + c_both_counts+=x[2] + else: + c_exception+=1 + c_exception_counts += x[2] + + + for x in t_samples: + + if "/" not in x[0]: + if len(x[0].split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + elif 0 == int(x[0].split("_")[3]): + t_5+=1 + t_5_counts += x[2] + elif 0 == int(x[0].split("_")[4]): + t_3+=1 + t_3_counts += x[2] + else: + t_both+=1 + t_both_counts+=x[2] + + else: + f=0 + for y in x[0].split("/"): + if len(y.split("_"))==2: + t_mature+=1 + t_mat_counts += x[2] + f=1 + break + if f==0: + part=x[0].split("/")[0].split("_")[-2]+"_"+x[0].split("/")[0].split("_")[-1] + num=0 + for y in x[0].split("/"): + if y.split("_")[-2]+"_"+y.split("_")[-1]==part: + num+=1 + if num==len(x[0].split("/")): + + if 0 == int(x[0].split("/")[0].split("_")[3]): + t_5+=1 + t_5_counts += x[2] + elif 0 == int(x[0].split("/")[0].split("_")[4]): + t_3+=1 + t_3_counts += x[2] + else: + t_both+=1 + t_both_counts+=x[2] + else: + t_exception+=1 + t_exception_counts += x[2] + + + c_all = c_5+c_3+c_both+c_mature+c_exception + c_all_counts = c_5_counts + c_3_counts + c_both_counts + c_mat_counts + c_exception_counts + + t_all = t_5+t_3+t_both+t_mature + t_exception + t_all_counts = t_5_counts + t_3_counts + t_both_counts + t_mat_counts + t_exception_counts + + c_5 = round(c_5/c_all*100,2) + c_3 = round(c_3/c_all*100,2) + c_both = round(c_both/c_all*100,2) + c_mature = round(c_mature/c_all*100,2) + c_exception = round(c_exception/c_all*100,2) + + c_5_counts = round(c_5_counts/c_all_counts*100,2) + c_3_counts = round(c_3_counts/c_all_counts*100,2) + c_both_counts = round(c_both_counts/c_all_counts*100,2) + c_mat_counts = round(c_mat_counts/c_all_counts*100,2) + c_exception_counts = round(c_exception_counts/c_all_counts*100,2) + + t_5 = round(t_5/t_all*100,2) + t_3 = round(t_3/t_all*100,2) + t_both = round(t_both/t_all*100,2) + t_mature = round(t_mature/t_all*100,2) + t_exception = round(t_exception/t_all*100,2) + + t_5_counts = round(t_5_counts/t_all_counts*100,2) + t_3_counts = round(t_3_counts/t_all_counts*100,2) + t_both_counts = round(t_both_counts/t_all_counts*100,2) + t_mat_counts = round(t_mat_counts/t_all_counts*100,2) + t_exception_counts = round(t_exception_counts/t_all_counts*100,2) + + radar_max = max(c_5, c_3, c_both,c_mature,c_exception,t_5,t_3,t_both,t_mature,t_exception) + radar_max_counts = max(c_5_counts,c_3_counts,c_both_counts,c_mat_counts,c_exception_counts,t_5_counts,t_3_counts,t_both_counts,t_mat_counts,t_exception_counts) + + df=pd.DataFrame({ + 'group':[group_name1,group_name2], + """5'3'-isomiRs""":[c_both,t_both], + """3'-isomiRs""":[c_3,t_3], + 'RefSeq miRNA':[c_mature,t_mature], + """5'-isomiRs""":[c_5,t_5], + 'Others*':[c_exception,t_exception]}) + + df1=pd.DataFrame({ + 'group':[group_name1,group_name2], + """5'-3'-isomiRs""":[c_both_counts,t_both_counts], + """3'-isomiRs""":[c_3_counts,t_3_counts], + 'RefSeq miRNA':[c_mat_counts,t_mat_counts], + """5'-isomiRs""":[c_5_counts,t_5_counts], + 'Others*':[c_exception_counts,t_exception_counts]}) + + spider_last(df,radar_max,1,group_name1,group_name2) + spider_last(df1,radar_max_counts,2,group_name1,group_name2) + + +############################################################################################################################################################################################################## + +def spider_last(df,radar_max,flag,group_name1,group_name2): + # ------- PART 1: Create background + fig = plt.figure() + # number of variable + categories=list(df)[1:] + N = len(categories) + + # What will be the angle of each axis in the plot? (we divide the plot / number of variable) + angles = [n / float(N) * 2 * pi for n in range(N)] + angles += angles[:1] + + # Initialise the spider plot + ax = plt.subplot(111, polar=True) + + # If you want the first axis to be on top: + ax.set_theta_offset(pi/2) + ax.set_theta_direction(-1) + + # Draw one axe per variable + add labels labels yet + plt.xticks(angles[:-1], categories, fontsize=11) + + # Draw ylabels + radar_max=round(radar_max+radar_max*0.1) + mul=len(str(radar_max))-1 + maxi=int(math.ceil(radar_max / pow(10,mul))) * pow(10,mul) + sep = round(maxi/4) + plt.yticks([sep, 2*sep, 3*sep, 4*sep, 5*sep], [str(sep)+'%', str(2*sep)+'%', str(3*sep)+'%', str(4*sep)+'%', str(5*sep)+'%'], color="grey", size=10) + plt.ylim(0, maxi) + + # ------- PART 2: Add plots + + # Plot each individual = each line of the data + # I don't do a loop, because plotting more than 3 groups makes the chart unreadable + + # Ind1 + values=df.loc[0].drop('group').values.flatten().tolist() + values += values[:1] + ax.plot(angles, values,'-o', linewidth=1, linestyle='solid', label=group_name1) + ax.fill(angles, values, 'b', alpha=0.1) + + # Ind2 + values=df.loc[1].drop('group').values.flatten().tolist() + values += values[:1] + ax.plot(angles, values, '-o' ,linewidth=1, linestyle='solid', label=group_name2) + ax.fill(angles, values, 'r', alpha=0.1) + + # Add legend + if flag==1: + plt.legend(loc='upper right', bbox_to_anchor=(0.0, 0.1)) + plt.savefig('spider_non_red.png',dpi=300) + else: + plt.legend(loc='upper right', bbox_to_anchor=(0.0, 0.1)) + plt.savefig('spider_red.png',dpi=300) + +############################################################################################################################################################################################################# + +def hist_red(samples,flag,group): + + lengths=[] + cat=[] + total_reads=0 + seq=[] + + if flag == "c": + title = "Length Distribution of "+group+" group (Redudant reads)" + if flag == "t": + title = "Length Distribution of "+group+" group (Redudant reads)" + + for i in samples: + for x in i: + lengths.append(x[3]) + if x[1]=="0": + seq.append([x[3],x[0].split("-")[1],"Mapped"]) + cat.append("Mapped") + if x[1] == "4": + seq.append([x[3],x[0].split("-")[1],"Unassigned"]) + cat.append("Unassigned") + + uni_len=list(set(lengths)) + uni_len=[x for x in uni_len if x<=35] + low=min(lengths) + up=max(lengths) + seq.sort() + uni_seq=list(seq for seq,_ in itertools.groupby(seq)) + dim=up-low + + if dim>20: + s=5 + else: + s=8 + + total_reads+=sum([int(x[1]) for x in uni_seq]) + + map_reads=[] + unmap_reads=[] + length=[] + for y in uni_len: + map_temp=0 + unmap_temp=0 + for x in uni_seq: + if x[0]==y and x[2]=="Mapped": + map_temp+=int(x[1]) + if x[0]==y and x[2]=="Unassigned": + unmap_temp+=int(x[1]) + if y<=35: + length.append(y) + map_reads.append(round(map_temp/total_reads*100,2)) + unmap_reads.append(round(unmap_temp/total_reads*100,2)) + + ylim=max([sum(x) for x in zip(unmap_reads, map_reads)]) + ylim=ylim+ylim*20/100 + fig, ax = plt.subplots() + width=0.8 + ax.bar(length, unmap_reads, width, label='Unassigned') + h=ax.bar(length, map_reads, width, bottom=unmap_reads, label='Mapped') + plt.xticks(np.arange(length[0], length[-1]+1, 1)) + plt.yticks(np.arange(0, ylim, 5)) + plt.xlabel('Length (nt)',fontsize=14) + plt.ylabel('Percentage',fontsize=14) + plt.title(title,fontsize=14) + ax.legend() + plt.ylim([0, ylim]) + ax.grid(axis='y',linewidth=0.2) + + if flag=='c': + plt.savefig('c_hist_red.png',dpi=300) + + if flag=='t': + plt.savefig('t_hist_red.png',dpi=300) + +############################################################################################################################################################################################################################################## + +def logo_seq_red(merge, flag, group): + + if flag=="c": + titlos = group + " group (Redundant)" + file_logo = "c_logo.png" + file_bar = "c_bar.png" + if flag=="t": + titlos = group + " group (Redundant)" + file_logo = "t_logo.png" + file_bar = "t_bar.png" + + c_samples=[[x[0],x[1],sum(int(i) for i in x[2:])] for x in merge] + + A=[0]*3 + C=[0]*3 + G=[0]*3 + T=[0]*3 + total_reads=0 + + for y in c_samples: + if "/" in y[0]: + length=[] + for x in y[0].split("/"): + length.append([len(x.split("_")[-1]),x.split("_")[-1],y[2]]) + + best=min(length) + total_reads+=best[2] + for i in range(3): + if i