import streamlit as st import pandas as pd import numpy as np from st_aggrid import AgGrid, GridOptionsBuilder,GridUpdateMode,DataReturnMode from iteration_utilities import duplicates from iteration_utilities import unique_everseen import os st.set_page_config(layout="wide") st.markdown( """ """, unsafe_allow_html=True, ) caution = '

Please note that Only one Guide (from pair) is found. Please see guides not found section for other guide

' caution1 = '

Please note that Each mutated guide is reported as a sepearte line. sgID_1/2, sgRNA_1/2, chr_sgRNA_1/2 and position_sgRNA_1/2 represent values for reference/mutated guide

' caution2 = '

Please Select a single/multiple guides and then select Check Box A, B or C Otherwise code will through error

' table_edit = '

About Table: Please note that table can be sorted by clicking on any column and Multiple rows can be selected (by clicking check box in first column) to save only those rows.

' caution_genes = '

Please make sure that desired genes from all three lists should be selected to generate Order Ready Table.

' #READ INPUT FILES cwd=os.getcwd()+'/'+'data/' #Here, gene column is modified for non-targeting guides in the format sgID_1|sgID_2 for coherent downstream manipulation listA = pd.read_csv(cwd+"guides_a_new.csv",index_col=False) listB = pd.read_csv(cwd+"guides_b_new.csv",index_col=False) listC = pd.read_csv(cwd+"guides_c_new.csv",index_col=False) lista_sz=listA.shape[0] listb_sz=listB.shape[0] listc_sz=listC.shape[0] #st.write(listA.shape) variantsa1=listA['gene'].unique() variantsb1=listB['gene'].unique() variantsc1=listC['gene'].unique() #Make a comprehensive lsit of genes in all 3 lists (Please not that non-targeting guide names are not same across three lists) con = np.concatenate((variantsa1, variantsb1, variantsc1)) variants_s=sorted(np.unique(con)) #NOW read GRCh38 and LR guides for stea as identified by LR-Guides pipeline #Format is: gene (as many entries as number of guides found, both matched and mutated), ref_guide, chr, position, mutated_guide (can also be same as reference), strand, num_mismatcg (excluding leading G), Please note that each guide has trailing NGG listA_found_ref = pd.read_csv(cwd+"seta_found_ref1.csv",index_col=False) listA_found_ref = listA_found_ref.sort_values('gene') lsita_ref_found_sz=listA_found_ref.shape[0] #remove # from chr# # listA_found_ref['chr'] = [x.split(' ')[-0] for x in listA_found_ref['chr']] listA_found_ref.rename(columns = {'strnad':'strand'}, inplace = True) #Also change strnad to strand (was misspelled in LR-Guides pipeline) #This (all such) file has 2-columns (gene as given in sgID_1/2, ref_guide). listA_notfound_ref = pd.read_csv(cwd+"seta_notfound_ref1.csv",index_col=False) listA_notfound_ref=listA_notfound_ref.sort_values('gene') lsita_ref_notfound_sz=listA_notfound_ref.shape[0] #LR guides listA_found_lr = pd.read_csv(cwd+"seta_found_LR1.csv",index_col=False) listA_found_lr=listA_found_lr.sort_values('gene') lsita_lr_found_sz=listA_found_lr.shape[0] listA_found_lr.rename(columns = {'strnad':'strand'}, inplace = True) listA_notfound_lr = pd.read_csv(cwd+"seta_notfound_LR1.csv",index_col=False) listA_notfound_lr=listA_notfound_lr.sort_values('gene') lsita_lr_notfound_sz=listA_notfound_lr.shape[0] #Also read GRCh38 and LR guides for set b listB_found_ref = pd.read_csv(cwd+"setb_found_ref1.csv",index_col=False) listB_found_ref=listB_found_ref.sort_values('gene') lsitb_ref_found_sz=listB_found_ref.shape[0] #remove # from chr# # listB_found_ref['chr'] = [x.split(' ')[-0] for x in listB_found_ref['chr']] listB_found_ref=listB_found_ref.sort_values('gene') listB_found_ref.rename(columns = {'strnad':'strand'}, inplace = True) listB_notfound_ref = pd.read_csv(cwd+"setb_notfound_ref1.csv",index_col=False) listB_notfound_ref=listB_notfound_ref.sort_values('gene') lsitb_ref_notfound_sz=listB_notfound_ref.shape[0] listB_found_lr = pd.read_csv(cwd+"setb_found_LR1.csv",index_col=False) listB_found_lr=listB_found_lr.sort_values('gene') lsitb_lr_found_sz=listB_found_lr.shape[0] listB_found_lr.rename(columns = {'strnad':'strand'}, inplace = True) listB_notfound_lr = pd.read_csv(cwd+"setb_notfound_LR1.csv",index_col=False) listB_notfound_lr=listB_notfound_lr.sort_values('gene') lsitb_lr_notfound_sz=listB_notfound_lr.shape[0] #Also read GRCh38 and LR guides for set c listC_found_ref = pd.read_csv(cwd+"setc_found_ref1.csv",index_col=False) listC_found_ref=listC_found_ref.sort_values('gene') lsitc_ref_found_sz=listC_found_ref.shape[0] #remove # from chr# # listC_found_ref['chr'] = [x.split(' ')[-0] for x in listC_found_ref['chr']] listC_found_ref.rename(columns = {'strnad':'strand'}, inplace = True) listC_notfound_ref = pd.read_csv(cwd+"setc_notfound_ref1.csv",index_col=False) listC_notfound_ref=listC_notfound_ref.sort_values('gene') lsitc_ref_notfound_sz=listC_notfound_ref.shape[0] listC_found_lr = pd.read_csv(cwd+"setc_found_LR1.csv",index_col=False) listC_found_lr=listC_found_lr.sort_values('gene') lsitc_lr_found_sz=listC_found_lr.shape[0] listC_found_lr.rename(columns = {'strnad':'strand'}, inplace = True) listC_notfound_lr = pd.read_csv(cwd+"setc_notfound_LR1.csv",index_col=False) listC_notfound_lr=listC_notfound_lr.sort_values('gene') lsitc_lr_notfound_sz=listC_notfound_lr.shape[0] #This for all guides order table set_start=0 regular_lista=listA[~listA['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] regular_lista=regular_lista.sort_values() set_end=regular_lista.shape[0] #18905 #regular_lista=regular_lista.iloc[set_start:set_end] non_targeting_lista=listA[listA['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] non_targeting_lista=non_targeting_lista.sort_values() #regular_lista=regular_lista.reset_index() regular_listb=listB[~listB['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] regular_listb=regular_listb.sort_values() #regular_listb=regular_listb.iloc[set_start:set_end] non_targeting_listb=listB[listB['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] non_targeting_listb=non_targeting_listb.sort_values() #regular_listb=regular_listb.reset_index() regular_listc=listC[~listC['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] regular_listc=regular_listc.sort_values() #regular_listc=regular_listc[set_start:set_end] non_targeting_listc=listC[listC['gene'].str.contains('non-targeting')]['sgID_AB']#[['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] non_targeting_listc=non_targeting_listc.sort_values() #GENERAL FUNCTIONS def transform(df,str): cols = st.multiselect(str, df.columns.tolist(), df.columns.tolist() ) df = df[cols] return df def convert_df(df): return df.to_csv().encode('utf-8') def convert_df1(df): return df.to_csv(index=False).encode('utf-8') #########TABLE DISPLAY def tbl_disp(dat,var,ref,key,flg=1): dat.reset_index(drop=True, inplace=True) #df = transform(dft,'Please Select columns to save whole table') #fname = st.text_input('Please input file name to save Table', 'temp') #fname = st_keyup("Please input file name to save Table", value='temp') csv = convert_df(dat) if flg==1: st.download_button( label="Download Full Table as CSV file", data=csv, file_name=var+'_'+ref+'.csv',#fname+'.csv', mime='text/csv', #key=key, ) gb = GridOptionsBuilder.from_dataframe(dat) gb.configure_pagination(enabled=False)#,paginationAutoPageSize=False)#True) #Add pagination gb.configure_default_column(enablePivot=True, enableValue=True, enableRowGroup=True) gb.configure_selection(selection_mode="multiple", use_checkbox=True) gb.configure_column("gene", headerCheckboxSelection = True) gb.configure_side_bar() gridOptions = gb.build() grid_response = AgGrid( dat, height=200, gridOptions=gridOptions, enable_enterprise_modules=True, update_mode=GridUpdateMode.MODEL_CHANGED, data_return_mode=DataReturnMode.FILTERED_AND_SORTED, fit_columns_on_grid_load=False, header_checkbox_selection_filtered_only=True, use_checkbox=True, width='100%' #key=key ) selected = grid_response['selected_rows'] if selected: #st.write('Selected rows') dfs = pd.DataFrame(selected) #st.dataframe(dfs[dfs.columns[1:dfs.shape[1]]]) #dfs1 = transform(dfs[dfs.columns[1:dfs.shape[1]]],'Please select columns to save selected Table') csv = convert_df1(dfs[dfs.columns[1:dfs.shape[1]]]) #csv = convert_df1(dfs1) if flg: st.download_button( label="Download Selected data as CSV", data=csv, file_name=var+'_'+ref+'.csv', mime='text/csv', ) return dfs def get_lists(ref_list,list_found_ref,list_notfound_ref): #This module retrieves guide_id and searches for guide sequences from the table #st.table(ref_list) a_ref=[] #st.table(ref_list) for i in range(len(ref_list)): a_ref.append(ref_list.sgID_AB.values[i].split('|')[0]) a_ref.append(ref_list.sgID_AB.values[i].split('|')[1]) set_found0_ref=[] #st.table(a_ref) for i in range(len(a_ref)): set_found0_ref.append(list_found_ref[list_found_ref['gene']==a_ref[i]]) #st.write(set_found0_ref) list_concatenated_found_ref = pd.concat(set_found0_ref) list_concatenated_match_ref = list_concatenated_found_ref[list_concatenated_found_ref.num_mismatch == 0] #only select guides with zero mismatches for match list, MISSMATCH LIST LATER #Also remove Alternate loci's data list_concatenated_match_ref = list_concatenated_match_ref[list_concatenated_match_ref['chr'].str.contains('chr')] #st.table(list_concatenated_match_ref) #also create new list with both sgRNAs in one row dft=pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) guideflg1=1 #st.table(list_concatenated_match_ref) if list_concatenated_match_ref.shape[0]>0: guideflg1=0 t=list_concatenated_match_ref.reset_index(drop=True) #st.table(t) ########## #check even/odd entries if t.shape[0]==1: t1=t.loc[t.index.repeat(2)].reset_index(drop=True) #st.write(t1) dft=assemble_tbl(t1) elif t.shape[0]%2==0: #even dft=assemble_tbl(t) else: #odd t1 = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) i=0 while i 0] list_concatenated_mutated_ref=list_concatenated_mutated_ref.sort_values('position') #Also remove Alternate loci's data list_concatenated_mutated_ref = list_concatenated_mutated_ref[list_concatenated_mutated_ref['chr'].str.contains('chr')] dft_mut = pd.DataFrame(columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2', 'sgID_1_2']) if list_concatenated_mutated_ref.shape[0]>0: dft_mut = get_mutated_res(list_concatenated_mutated_ref) #check not found seta_notfound0_ref=list_notfound_ref[list_notfound_ref['gene']==a_ref[0]] seta_notfound1_ref=list_notfound_ref[list_notfound_ref['gene']==a_ref[1]] list_concatenated_notfound_ref = pd.concat([seta_notfound0_ref,seta_notfound1_ref]) return dft.iloc[:1], dft_mut,list_concatenated_notfound_ref,list_concatenated_match_ref,list_concatenated_mutated_ref,guideflg1 ########### def get_mutated_res(list_concatenated_mutated_ref): ######### #if list_concatenated_mutated_ref.shape[0]>0: t=list_concatenated_mutated_ref.reset_index(drop=True) #st.table(t) dft_mut = pd.DataFrame(columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2', 'sgID_1_2']) c1=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1'] c2=['sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2']#, 'sgID_1_2'] #st.table(listA_concatenated_match_ref) #st.write(t.shape[0]) tf=0 #for i in range(0,t.shape[0],2): for i in range(t.shape[0]): l1=t.iloc[[i]] l1.columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','mutated_guide', 'strand', 'num_mismatch'] l2=l1.copy() l2.columns=['sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2','mutated_guide2', 'strand2', 'num_mismatch2'] list_concatenated_mutated_ref1=[] #listA_concatenated_mutated_ref1=pd.concat([l1.reset_index(drop=True),l2.reset_index(drop=True)],axis=1) list_concatenated_mutated_ref1=pd.concat([l1.reset_index(drop=True),l2.reset_index(drop=True)],axis=1) #st.table(listA_concatenated_mutated_ref1) list_concatenated_mutated_ref1=list_concatenated_mutated_ref1[['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','mutated_guide2','chr_sgRNA_2','position_sgRNA_2']] #also change if not leading G list_concatenated_mutated_ref1['sgRNA_1']='G'+list_concatenated_mutated_ref1['sgRNA_1'].str.slice(1, 20) #also change name of mutated_guide2 column list_concatenated_mutated_ref1.columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2'] list_concatenated_mutated_ref1['sgRNA_2']='G'+list_concatenated_mutated_ref1['sgRNA_2'].str.slice(1, 20) list_concatenated_mutated_ref1['sgID_1_2']=list_concatenated_mutated_ref1['sgID_1']+"|"+list_concatenated_mutated_ref1['sgID_1'] #dft_mut=dft_mut.append(list_concatenated_mutated_ref1) dft_mut=pd.concat([dft_mut,list_concatenated_mutated_ref1]) return dft_mut def not_found_check(set12,set34,set56,listA_notfound_lr,listB_notfound_lr,listC_notfound_lr): flg11=0 flg12=0 flg21=0 flg22=0 flg31=0 flg32=0 #st.write(set12.split('|')[1]) if listA_notfound_lr[listA_notfound_lr['gene']==set12.split('|')[0]].shape[0]>0: flg11=1 if listA_notfound_lr[listA_notfound_lr['gene']==set12.split('|')[1]].shape[0]>0: flg12=1 if listB_notfound_lr[listB_notfound_lr['gene']==set34.split('|')[0]].shape[0]>0: flg21=1 if listB_notfound_lr[listB_notfound_lr['gene']==set34.split('|')[1]].shape[0]>0: flg22=1 if listC_notfound_lr[listC_notfound_lr['gene']==set56.split('|')[0]].shape[0]>0: flg31=1 if listC_notfound_lr[listC_notfound_lr['gene']==set56.split('|')[1]].shape[0]>0: flg32=1 return flg11,flg12,flg21,flg22,flg31,flg32 def order_ready_tbl_CHM13(set12,set34,set56,listA_found_lr,listA_notfound_lr,listB_found_lr,listB_notfound_lr,listC_found_lr,listC_notfound_lr,ref_sel): # st.table(set12) # st.table(set34) # st.table(set56) dft_order_table=pd.DataFrame(columns=['gene','guide_type','sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_notfound_all=pd.DataFrame(columns=['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']) #dft_notfound=pd.DataFrame(columns=['gene','ref_guide']) dft_a = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_b = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_c = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) set12=set12.reset_index(drop = True) set34=set34.reset_index(drop = True) set56=set56.reset_index(drop = True) for i in range(set12.shape[0]): gene_n=set12[i].split('_')[0] f=not_found_check(set12[i],set34[i],set56[i],listA_notfound_lr,listB_notfound_lr,listC_notfound_lr) #st.write(f) #st.write(set12[i],set34[i],set56[i]) #ref_listA=listA[listA['gene']==variant_set.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listA=listA[listA['sgID_AB']==set12.iloc[i]][['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listA = ref_listA[['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']] #st.write(ref_listA) #ref_listA.columns=['gene','guide_type','protospacer_A','protospacer_B'] resa,res_muta,res_notfounda,list_matcha,list_mutateda,gflga1=get_lists(ref_listA,listA_found_lr,listA_notfound_lr) #dft_a=dft_a.append(ref_listA) #listb ref_listB=listB[listB['sgID_AB']==set34.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listB = ref_listB[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listB.columns=['gene','guide_type','protospacer_A','protospacer_B'] resb,res_mutb,res_notfoundb,list_matchb,list_mutatedb,gflgb1=get_lists(ref_listB,listB_found_lr,listB_notfound_lr) #dft_b=dft_b.append(ref_listB) #listc ref_listC=listC[listC['sgID_AB']==set56.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listC = ref_listC[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listC.columns=['gene','guide_type','protospacer_A','protospacer_B'] resc,res_mutc,res_notfoundc,list_matchc,list_mutatedc,gflgc1=get_lists(ref_listC,listC_found_lr,listC_notfound_lr) #dft_c=dft_c.append(ref_listC) #st.table(ref_listA) # st.write(gflga1,gflgb1,gflgc1) if gflga1==0: #Also verigy that both guides are different #st.table(resa) if resa['sgID_1'][0] != resa['sgID_2'][0]: resa['gene']=gene_n resa['guide_type']='1-2' #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table, resa]) #dft_order_table.concat(resa) else: #it is nutation case, so check next if f[2]==0 or f[3] == 0: #st.write('came in 1') if not resb.empty: # and resb['sgID_1'][0] != resb['sgID_2'][0]: #second guide in from setb resa[['sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2']] = resb[['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1']] resa['sgID_1_2'] = resa['sgID_1']+"|"+resa['sgID_2'] if f[2]==0: resa['gene']=gene_n if f[0]==0: resa['guide_type']="1-3" else: resa['guide_type']="2-3" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: # f[2]==0: resa['gene']=gene_n if f[0]==0: resa['guide_type']="1-4" else: resa['guide_type']="2-4" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) elif resa.shape[0] >0: #at least one guide is from seta #if resa['sgID_1'][0] != resa['sgID_2'][0]: if f[2]==0 or f[3] == 0: #st.write('came in 1') if not resb.empty: # and resb['sgID_1'][0] != resb['sgID_2'][0]: #second guide in from setb resa[['sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2']] = resb[['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1']] resa['sgID_1_2'] = resa['sgID_1']+"|"+resa['sgID_2'] if f[2]==0: resa['gene']=gene_n resa['guide_type']=str(gflga1)+"-3" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: # f[2]==0: resa['gene']=gene_n resa['guide_type']=str(gflga1)+"-4" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) elif f[4]==0 or f[5] == 0: #st.write('came in 2') #if resa['sgID_1'][0] != resa['sgID_2'][0]: if not resc.empty: # and resc['sgID_1'][0] != resc['sgID_2'][0]: # resc.shape[0]>0: #second guide is from setc resa[['sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2']] = resc[['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1']] resa['sgID_1_2'] = resa['sgID_1']+"|"+resa['sgID_2'] #dft_order_table=dft_order_table.append(resa) if f[4]==0: resa['gene']=gene_n resa['guide_type']=str(gflga1)+"-5" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: # f[2]==0: resa['gene']=gene_n resa['guide_type']=str(gflga1)+"-6" #dft_order_table=dft_order_table.append(resa) dft_order_table=pd.concat([dft_order_table,resa]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) elif resb.shape[0]>0: #at least one guide if gflgb1==0: if resb['sgID_1'][0] != resb['sgID_2'][0]: resb['gene']=gene_n resb['guide_type']='3-4' #dft_order_table=dft_order_table.append(resb) dft_order_table=pd.concat([dft_order_table,resb]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) elif f[4]==0 or f[5] == 0: #if not resc.empty and resc['sgID_1'][0] != resc['sgID_2'][0]: resb[['sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2']] = resc[['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1']] resb['sgID_1_2'] = resb['sgID_1']+"|"+resb['sgID_2'] #dft_order_table=dft_order_table.append(resb) if f[4]==0: resb['gene']=gene_n resb['guide_type']=str(gflgb1+2)+"-5" #dft_order_table=dft_order_table.append(resb) dft_order_table=pd.concat([dft_order_table,resb]) else: # f[2]==0: resb['gene']=gene_n resb['guide_type']=str(gflgb1+2)+"-6" #dft_order_table=dft_order_table.append(resb) dft_order_table=pd.concat([dft_order_table,resb]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) elif resc.shape[0]>0: #at least one guide if gflgc1==0: if resc['sgID_1'][0] != resc['sgID_2'][0]: resc['gene']=gene_n resc['guide_type']='5-6' #dft_order_table=dft_order_table.append(resc) dft_order_table=pd.concat([dft_order_table,resc]) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) else: dft_notfound_all=pd.concat([dft_notfound_all,ref_listA], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listB], ignore_index = True) dft_notfound_all=pd.concat([dft_notfound_all,ref_listC], ignore_index = True) if dft_order_table.shape[0]>0: #check total guides found # st.write(str(set12.shape[0])) # st.write(str(set34.shape[0])) # st.write(str(set56.shape[0])) st.write('**Please note that for guides matching to multiple locations (an example is ABCC6), only first pair is returned**') szt=set12.shape[0] szf=dft_order_table.shape[0] # st.write(str(dft_order_table.shape[0])) szd=szt-szf if szd>0: st.write('Order Ready '+ref_sel+' guides List: '+str(szd)+'/'+str(szt)+' **guides were not found**') tbl_disp(dft_order_table,'select_genes','SetA_CHM13',5) else: st.write('Order Ready '+ref_sel+' guides List') tbl_disp(dft_order_table,'select_genes','SetA_CHM13',5) else: st.write('**No guides found in ListA, ListB and ListC**') if dft_notfound_all.shape[0]>0: st.write('**Guides not found in any lists**') tbl_disp(dft_notfound_all,'select_genes','SetA_CHM13',6) def assemble_tbl(t): dft = pd.DataFrame(columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2', 'sgID_1_2']) #for i in range(0,t.shape[0],2): mid=int(t.shape[0]/2) for i in range(int(t.shape[0]/2)): l1=t.iloc[[i]] l1.columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','mutated_guide', 'strand', 'num_mismatch'] #l2=t.iloc[[i+1]] l2=t.iloc[[mid]] l2.columns=['sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2','mutated_guide2', 'strand2', 'num_mismatch2'] listA_concatenated_match_LR1=pd.concat([l1.reset_index(drop=True),l2.reset_index(drop=True)],axis=1) listA_concatenated_match_LR1=listA_concatenated_match_LR1[['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2']] listA_concatenated_match_LR1['sgRNA_1']=listA_concatenated_match_LR1['sgRNA_1'].str.slice(0, 20) listA_concatenated_match_LR1['sgRNA_2']=listA_concatenated_match_LR1['sgRNA_2'].str.slice(0, 20) listA_concatenated_match_LR1['sgID_1_2']=listA_concatenated_match_LR1['sgID_1']+"|"+listA_concatenated_match_LR1['sgID_2'] #dft=dft.append(listA_concatenated_match_LR1) dft=pd.concat([dft,listA_concatenated_match_LR1]) mid=mid+1 return dft #Get non-targeting lists def get_lists_non_targeting(ref_list,list_found_ref,list_notfound_ref): #This module retrieves guide_id and searches for guide sequences from the table #st.table(ref_list) a_ref=[] for i in range(len(ref_list)): a_ref.append(ref_list.sgID_AB.values[i].split('|')[0]) a_ref.append(ref_list.sgID_AB.values[i].split('|')[1]) set_found0_ref=[] for i in range(len(a_ref)): set_found0_ref.append(list_found_ref[list_found_ref['gene']==a_ref[i]]) list_concatenated_found_ref = pd.concat(set_found0_ref) list_concatenated_match_ref = list_concatenated_found_ref[list_concatenated_found_ref.num_mismatch == 0] #only select guides with zero mismatches for match list, MISSMATCH LIST LATER #get matching to Alternating loci's list_concatenated_match_alt_ref = list_concatenated_match_ref[~list_concatenated_match_ref['chr'].str.contains('chr')] #Also remove Alternate loci's data list_concatenated_match_ref = list_concatenated_match_ref[list_concatenated_match_ref['chr'].str.contains('chr')] #st.table(list_concatenated_match_ref) #also create new list with both sgRNAs in one row dft=pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) if list_concatenated_match_ref.shape[0]>0: t=list_concatenated_match_ref.reset_index(drop=True) #st.table(t) ########## #check even/odd entries if t.shape[0]==1: t1=t.loc[t.index.repeat(2)].reset_index(drop=True) #st.write(t1) dft=assemble_tbl(t1) elif t.shape[0]%2==0: #even dft=assemble_tbl(t) else: #odd t1 = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) i=0 while i 0] list_concatenated_mutated_ref=list_concatenated_mutated_ref.sort_values('position') #Also remove Alternate loci's data list_concatenated_mutated_alt_ref = list_concatenated_mutated_ref[~list_concatenated_mutated_ref['chr'].str.contains('chr')] list_concatenated_mutated_ref = list_concatenated_mutated_ref[list_concatenated_mutated_ref['chr'].str.contains('chr')] dft_mut = pd.DataFrame(columns=['sgID_1','sgRNA_1','chr_sgRNA_1','position_sgRNA_1','sgID_2','sgRNA_2','chr_sgRNA_2','position_sgRNA_2', 'sgID_1_2']) if list_concatenated_mutated_ref.shape[0]>0: dft_mut = get_mutated_res(list_concatenated_mutated_ref) #check not found seta_notfound0_ref=list_notfound_ref[list_notfound_ref['gene']==a_ref[0]] seta_notfound1_ref=list_notfound_ref[list_notfound_ref['gene']==a_ref[1]] #st.write(list_notfound_ref[list_notfound_ref['gene']==a_ref[0]]) #st.write(seta_notfound0_ref) #st.write(seta_notfound1_ref) #add guideflg1 to return which guide is found guideflg1=0 if seta_notfound0_ref.shape[0]>0: guideflg1=2 if seta_notfound1_ref.shape[0]>0: guideflg1=1 list_concatenated_notfound_ref = pd.concat([seta_notfound0_ref,seta_notfound1_ref]) #st.table(a_ref) #st.table(seta_notfound1_ref) #st.table(dft) #st.table(dft_mut) return dft, dft_mut,list_concatenated_notfound_ref,list_concatenated_match_ref,list_concatenated_mutated_ref,list_concatenated_match_alt_ref,list_concatenated_mutated_alt_ref,guideflg1 ########### #Get All Guides Stats #def process_all_guides(glist,list,ref_type,guide_type): def process_all_guides(glist,for_list,f_list,nf_list): #st.write(type(glist)) #st.table(for_list) #for_list=for_list.reset_index() variant_set=glist['gene'] dft_c = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_resc=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_res_mutc=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_notfoundc=pd.DataFrame(columns=['gene','ref_guide']) df_matched_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_matched_alt_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_mutated_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_mutated_guides_alt_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) #st.table(for_list) for i in range(variant_set.shape[0]): #st.write(variant_set.iloc[i]) ref_listC=for_list[for_list['sgID_AB']==variant_set.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listC =ref_listC[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #st.table(ref_listC) #st.table(ref_listC) res,res_mut,res_notfound,list_match,list_mutated,list_match_alt,list_mutated_alt,gflgc1=get_lists_non_targeting(ref_listC,f_list,nf_list) #dft_c=dft_c.append(ref_listC) if res.shape[0]>0: dft_resc=pd.concat([dft_resc,res]) if res_mut.shape[0]>0: dft_res_mutc=pd.concat([dft_res_mutc,res_mut]) if res_notfound.shape[0]>0: dft_notfoundc= pd.concat([dft_notfoundc,res_notfound]) if list_match.shape[0]>0: df_matched_guides_ref= pd.concat([df_matched_guides_ref,list_match]) if list_mutated.shape[0]>0: df_mutated_guides_ref= pd.concat([df_mutated_guides_ref,list_mutated]) if list_match_alt.shape[0]>0: df_matched_alt_ref=pd.concat([df_matched_alt_ref,list_mutated]) if list_mutated_alt.shape[0]>0: df_mutated_guides_alt_ref=pd.concat([df_mutated_guides_alt_ref,list_mutated_alt]) if df_matched_guides_ref.shape[0]>0: #st.write(type(df_matched_guides_ref['gene'])) gl=df_matched_guides_ref['gene'] dupesm=gl[gl.duplicated()] if df_mutated_guides_ref.shape[0]>0: gl=df_mutated_guides_ref['gene'] dupesmu=gl[gl.duplicated()] #now check common between matched and mutated # if dupesm.shape[0]>0 and dupesmu.shape[0]>0: # common_list = set(dupesm).intersection(dupesmu) # st.table(common_list) # st.write('common guides between matched and mutated lists are: '+len(common_list)) if df_matched_guides_ref.shape[0]>0: if dupesm.shape[0]>0: st.write('**Matched Guides**: '+str(df_matched_guides_ref.shape[0])+' and: '+str(dupesm.shape[0])+' are repeated guides (matched to multiple locations)') tbl_disp(df_matched_guides_ref,'select_genes','SetC_GRCh38',17) #st.table(dupesm,'select_genes','SetC_GRCh38',17) tbl_disp(dupesm,'select_genes','SetC_GRCh38',17) else: st.write('**Matched Guides**: '+str(df_matched_guides_ref.shape[0])) tbl_disp(df_matched_guides_ref,'select_genes','SetC_GRCh38',17) if df_matched_alt_ref.shape[0]>0: st.write('**Matched Guides to Alt Loci**: '+str(df_matched_alt_ref.shape[0])) tbl_disp(df_matched_alt_ref,'select_genes','SetC_GRCh38',17) if df_mutated_guides_ref.shape[0]>0: #gl=df_mutated_guides_ref['gene'] #dupesmu=gl[gl.duplicated()] if dupesmu.shape[0]>0: st.write('**Mutated Guides (some might have >1 guides)**: '+str(df_mutated_guides_ref.shape[0])+' and: '+str(dupesmu.shape[0])+' are repeated guides') tbl_disp(df_mutated_guides_ref,'select_genes','SetC_GRCh38',18) #st.table(dupesmu) else: st.write('**Mutated Guides (some might have >1 guides)**: '+str(df_mutated_guides_ref.shape[0])) tbl_disp(df_mutated_guides_ref,'select_genes','SetC_GRCh38',18) if df_mutated_guides_alt_ref.shape[0]>0: st.write('**Mutated Guides to Alt Loci**: '+str(df_mutated_guides_alt_ref.shape[0])) tbl_disp(df_mutated_guides_alt_ref,'select_genes','SetC_GRCh38',18) if dft_notfoundc.shape[0]>0: st.write('**Guides Not Found**: '+str(dft_notfoundc.shape[0])) tbl_disp(dft_notfoundc,'select_genes','SetC_GRCh38',19) #CALC BASED ON LIST, GUIDE TYPE AND REFERENCE #END GENERAL FUNCTIONS st.title('Long Read Guides Search') st.write('**Important:** Please note that **MTMR3** is not present in guides_c list, so we have **removed it from list a and list b**') #tbl_disp(regulara,'variant','ref_guides',0,1) Calc = st.sidebar.radio( "", ('ReadME', 'Single/Multiple Guides','All','Not_Found')) if Calc == 'ReadME': expander = st.expander("How to use this app") #st.header('How to use this app') expander.markdown('Please select **Single Gene** OR **Multiple Genes** Menue checkbox from the sidebar') expander.markdown('Select a Gene (from genes dropdown list) OR Multiple genes (from table)') expander.markdown('A table showing all reference gudies from three LISTS will appear in the main panel. **Please not some of the genes (for example A1BG and GJB7) have multiple guide pairs and all of these are selected.**') expander.markdown('To see results for each of the selected reference guide from ListA, ListB and ListC, Please select respective checkbox') expander.markdown('Results are shown as two tables, **Matched** and **Mutated** guides tables and **NOT FOUND** table if guides are not found in GRCh38 and LR reference fasta files') expander.markdown('**Mutated** guides table shows the genomic postion in GRCh38 and LR Fasta file along other fields. **If a guide is found in GRCh38 but not in LR fasta, then corresponding columns will be NA**') expander.markdown('**Mutated** guides table shows the genomic postion in GRCh38 and LR Fasta file along other fields. **If a guide is found in GRCh38 but not in LR fasta, then corresponding columns will be NA**') expander1 = st.expander('Introduction') expander1.markdown( """ This app helps navigate all probable genomic **miss-matched/Mutations (upto 2 bp)** for a given sgRNA (from 3 lists of CRISPRi dual sgRNA libraries) in GRCh38 reference fasta and a Reference fasta generated from BAM generated against KOLF2.1J longread data. """ ) expander1.markdown('Merged bam file was converted to fasta file using following steps:') expander1.markdown('- samtools mpileup to generate bcf file') expander1.markdown('- bcftools to generate vcf file') expander1.markdown('- bcftools consensus to generate fasta file') expander1.markdown('A GPU based [Cas-OFFinder](http://www.rgenome.net/cas-offinder/) tool was used to find off-target sequences (upto 2 miss-matched) for each geiven reference guide against GRCh38 and LR fasta references.') elif Calc=='Single/Multiple Guides': flg_a_fount=0 flg_b_fount=0 flg_c_fount=0 #st.write('**General Stats:**') #st.write('**GRCh38 Stats: Guides Found: **'+str(lsita_ref_found_sz)+"/"+str(lista_sz)) with st.form(key='columns_in_form'): c2, c3 = st.columns(2) with c2: multi_genes = st.multiselect( 'Please select genes list to start processing', variants_s) Updated=st.form_submit_button(label = 'Update') listA_concatenated_orig = pd.DataFrame(columns=['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']) reflistA_concatenated = pd.DataFrame(columns=['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']) reflistB_concatenated = pd.DataFrame(columns=['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']) reflistC_concatenated = pd.DataFrame(columns=['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']) for variant in multi_genes: ref_listA=listA[listA['gene']==variant][['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listA = ref_listA[['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listA.columns=['gene','guide_type','protospacer_A','protospacer_B'] reflistA_concatenated=pd.concat([reflistA_concatenated,ref_listA]) ref_listB=listB[listB['gene']==variant][['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listB = ref_listB[['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listB.columns=['gene','guide_type','protospacer_A','protospacer_B'] reflistB_concatenated=pd.concat([reflistB_concatenated,ref_listB]) ref_listC=listC[listC['gene']==variant][['gene','guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listC = ref_listC[['gene','sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listC.columns=['gene','guide_type','protospacer_A','protospacer_B'] reflistC_concatenated=pd.concat([reflistC_concatenated,ref_listC]) listA_concatenated_orig = pd.concat([listA_concatenated_orig,ref_listA,ref_listB,ref_listC]) if listA_concatenated_orig.shape[0] > 0: #st.markdown(table_edit,unsafe_allow_html=True) st.write('**Input** Guides (all 6 from 3 sets).') st.write('**Please Select Guides common to ALL 3 Lists to procede further Processing**') st.markdown(caution_genes,unsafe_allow_html=True) with st.form(key='columns_in_form_a'): c2, c3 = st.columns([10,2]) with c2: get_table_order=tbl_disp(listA_concatenated_orig,'variant','ref_guides',111,0) with c3: ref_sel = st.radio("Select Reference", ('CHM13','GRCh38'), horizontal=True) Updated1=st.form_submit_button(label = 'Generate Order Ready Table') if not isinstance(get_table_order, type(None)): # and Updated1:# and get_table_order.shape[0]>0: if ref_sel=='GRCh38': list_founda=listA_found_ref list_notfounda=listA_notfound_ref list_foundb=listB_found_ref list_notfoundb=listB_notfound_ref list_foundc=listC_found_ref list_notfoundc=listC_notfound_ref else: list_founda=listA_found_lr list_notfounda=listA_notfound_lr list_foundb=listB_found_lr list_notfoundb=listB_notfound_lr list_foundc=listC_found_lr list_notfoundc=listC_notfound_lr variant_set12=get_table_order[get_table_order['guide_type']=='1-2']['sgID_AB'] variant_set34=get_table_order[get_table_order['guide_type']=='3-4']['sgID_AB'] variant_set56=get_table_order[get_table_order['guide_type']=='5-6']['sgID_AB'] #st.table(variant_set12) #st.write(variant_set12) if variant_set12.shape[0]==variant_set34.shape[0]==variant_set56.shape[0]: #########Here we call order ready table #order_ready_tbl_GRCh38(variant_set12,variant_set34,variant_set56) #order_ready_tbl_CHM13(variant_set12,variant_set34,variant_set56,listA_found_lr,listA_notfound_lr,listB_found_lr,listB_notfound_lr,listC_found_lr,listC_notfound_lr) order_ready_tbl_CHM13(variant_set12,variant_set34,variant_set56,list_founda,list_notfounda,list_foundb,list_notfoundb,list_foundc,list_notfoundc,ref_sel) ########END ORDER READY TABLE elif variant_set12.shape[0]!=variant_set34.shape[0]: st.markdown("""**SetA and SetB guides are not same, Please correct the problem and re-run**""",unsafe_allow_html=True) elif variant_set12.shape[0]!=variant_set56.shape[0]: st.markdown("""**SetA and SetC guides are not same, Please correct the problem and re-run**""",unsafe_allow_html=True) elif variant_set34.shape[0]!=variant_set56.shape[0]: st.markdown("""**SetB and SetC guides are not same, Please correct the problem and re-run**""",unsafe_allow_html=True) else: st.markdown("""**Probably Mixed guides are selected from three lists, Please correct the problem and re-run**""",unsafe_allow_html=True) else: st.write('**Please select guides and Press Update Button to Begin Processing**') if 'get_table_order' in locals(): if not isinstance(get_table_order, type(None)): st.write('**For List wise results, Please select a List**') reflistA_concatenated=get_table_order[get_table_order['guide_type']=='1-2'] reflistA_concatenated.drop("_selectedRowNodeInfo",axis=1,inplace=True) reflistB_concatenated=get_table_order[get_table_order['guide_type']=='3-4'] reflistB_concatenated.drop("_selectedRowNodeInfo",axis=1,inplace=True) reflistC_concatenated=get_table_order[get_table_order['guide_type']=='5-6'] reflistC_concatenated.drop("_selectedRowNodeInfo",axis=1,inplace=True) #st.write('**Important:** If a guides is **not** in **found, mutated and not_found list (such as GSTT1), then it is found in Alternative Loci and Removed**') with st.form(key='columns_in_form_lists'): c2, c3= st.columns([10,1])#([10,10]) with c2: List_Selected = st.selectbox('Please select list', ('','ListA','ListB','ListC')) Show_ListResults=st.form_submit_button(label = 'GO') #ListARes = st.checkbox('Results For SetA',key=300) if List_Selected=='ListA':# and not isinstance(get_table, type(None)):#get_table!=None: ref_list= listA st.write('**Please select Guides From Table Below to processes from ListA**') with st.form(key='columns_in_form_listsA'): c2, c3= st.columns([100,2])#([10,10]) with c2: get_table=tbl_disp(reflistA_concatenated,variant,'ref_guides',2,0) #List_Selected = st.selectbox('Please select list', #('ListA','ListB','ListC')) Show_ListResults=st.form_submit_button(label = 'Show ListA Results') #st.write('**Please select Guides From Table Below to processes from ListA**') #get_table=tbl_disp(reflistA_concatenated,variant,'ref_guides',2,0) if not isinstance(get_table, type(None)): if ref_sel=='GRCh38': list_found=listA_found_ref list_notfound=listA_notfound_ref else: list_found=listA_found_lr list_notfound=listA_notfound_lr variant_set=get_table['sgID_AB'] dft_a = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_resa=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_res_muta=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_notfounda=pd.DataFrame(columns=['gene','ref_guide']) df_matched_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_mutated_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) #CHECK FOR GRCh38 for i in range(variant_set.shape[0]): #ref_listA=listA[listA['sgID_AB']==variant_set.iloc[i]['gene']][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listA=ref_list[ref_list['sgID_AB']==variant_set.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listA = ref_listA[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listA.columns=['gene','guide_type','protospacer_A','protospacer_B'] #st.table(ref_listA) res,res_mut,res_notfound,list_match,list_mutated,gflga1=get_lists(ref_listA,list_found,list_notfound) #dft_a=dft_a.append(ref_listA) if res.shape[0]>0: dft_resa=pd.concat([dft_resa,res]) if res_mut.shape[0]>0: dft_res_muta=pd.concat([dft_res_muta,res_mut]) if res_notfound.shape[0]>0: dft_notfounda= pd.concat([dft_notfounda,res_notfound]) if list_match.shape[0]>0: df_matched_guides_ref= pd.concat([df_matched_guides_ref,list_match]) if list_mutated.shape[0]>0: df_mutated_guides_ref= pd.concat([df_mutated_guides_ref,list_mutated]) #st.write('Selected Reference Guides for **Set A**') #tbl_disp(dft_a,'All','ReferenceGuides',0) st.write('**Important:** If a guides is **not** in **found, mutated and not_found list (such as GSTT1), then it is found in Alternative Loci and Removed**') if dft_resa.shape[0]>0: st.write('Matched to '+ref_sel+' Reference Guides for **Set A**') tbl_disp(dft_resa,'select_genes','SetA_GRCh38',3) elif dft_res_muta.shape[0]>0: st.write('None of the guides Matched, So reporting **Mutated to** '+ref_sel+' Reference Guides for **Set A**') st.markdown(caution1,unsafe_allow_html=True) tbl_disp(dft_res_muta,'select_genes','SetA_Mutated_GRCh38',4) if dft_notfounda.shape[0]>0: st.write('**SetA Guides Not Found in '+ref_sel+' (None of the guides are Matched/Mutated)**') #tbl_disp(dft_notfound,'select_genes','SetA_Notfound_GRCh38') st.table(dft_notfounda) #ListBRes = st.checkbox('Results For SetB',key=40) if List_Selected=='ListB': # and not isinstance(get_table, type(None)):#get_table!=None: ref_list= listB st.write('**Please select Guides From Table Below to processes from ListB**') with st.form(key='columns_in_form_listsA'): c2, c3= st.columns([100,2])#([10,10]) with c2: get_table=tbl_disp(reflistB_concatenated,variant,'ref_guides',2,0) Show_ListResults=st.form_submit_button(label = 'Show ListB Results') if not isinstance(get_table, type(None)): if ref_sel=='GRCh38': list_found=listB_found_ref list_notfound=listB_notfound_ref else: list_found=listB_found_lr list_notfound=listB_notfound_lr #variant_set=get_table[['gene']] variant_set=get_table['sgID_AB'] dft_b = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_resb=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_res_mutb=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_notfoundb=pd.DataFrame(columns=['gene','ref_guide']) df_matched_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_mutated_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) #CHECK FOR GRCh38 for i in range(variant_set.shape[0]): #ref_listB=listB[listB['gene']==variant_set.iloc[i]['gene']][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listB=ref_list[ref_list['sgID_AB']==variant_set.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listB =ref_listB[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listB.columns=['gene','guide_type','protospacer_A','protospacer_B'] res,res_mut,res_notfound,list_match,list_mutated,gflgb1=get_lists(ref_listB,list_found,list_notfound) #dft_b=dft_b.append(ref_listB) if res.shape[0]>0: dft_resb=pd.concat([dft_resb,res]) if res_mut.shape[0]>0: dft_res_mutb=pd.concat([dft_res_mutb,res_mut]) if res_notfound.shape[0]>0: dft_notfoundb= pd.concat([dft_notfoundb,res_notfound]) if list_match.shape[0]>0: df_matched_guides_ref= pd.concat([df_matched_guides_ref,list_match]) if list_mutated.shape[0]>0: df_mutated_guides_ref= pd.concat([df_mutated_guides_ref,list_mutated]) #st.write('Selected Reference Guides for **Set B**') #tbl_disp(dft_b,'All','ReferenceGuides',0) st.write('**Important:** If a guides is **not** in **found, mutated and not_found list (such as GSTT1), then it is found in Alternative Loci and Removed**') if dft_resb.shape[0]>0: st.write('Matched to '+ref_sel+' Reference Guides for **Set B**') tbl_disp(dft_resb,'select_genes','SetB_GRCh38',10) elif dft_res_mutb.shape[0]>0: st.write('None of the guides Matched, So reporting **Mutated to '+ref_sel+' Reference Guides for **Set B**') st.markdown(caution1,unsafe_allow_html=True) tbl_disp(dft_res_mutb,'select_genes','SetB_Mutated_GRCh38',11) if dft_notfoundb.shape[0]>0: st.write('**SetB Guides Not Found in '+ref_sel+' (None of the guides are Matched/Mutated)**') #tbl_disp(dft_notfound,'select_genes','SetA_Notfound_GRCh38') st.table(dft_notfoundb) #ListCRes = st.checkbox('Results For SetC',key=50) if List_Selected=='ListC': # and not isinstance(get_table, type(None)):#get_table!=None: ref_list= listC st.write('**Please select Guides From Table Below to processes from ListC**') with st.form(key='columns_in_form_listsA'): c2, c3= st.columns([100,2])#([10,10]) with c2: get_table=tbl_disp(reflistC_concatenated,variant,'ref_guides',2,0) Show_ListResults=st.form_submit_button(label = 'Show ListC Results') if not isinstance(get_table, type(None)): if ref_sel=='GRCh38': list_found=listC_found_ref list_notfound=listC_notfound_ref else: list_found=listC_found_lr list_notfound=listC_notfound_lr variant_set=get_table['sgID_AB'] dft_c = pd.DataFrame(columns=['gene','guide_type','protospacer_A','protospacer_B']) dft_resc=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_res_mutc=pd.DataFrame(columns=['sgID_1', 'sgRNA_1', 'chr_sgRNA_1', 'position_sgRNA_1', 'sgID_2', 'sgRNA_2', 'chr_sgRNA_2', 'position_sgRNA_2', 'sgID_1_2']) dft_notfoundc=pd.DataFrame(columns=['gene','ref_guide']) df_matched_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) df_mutated_guides_ref = pd.DataFrame(columns=['gene','ref_guide', 'chr', 'position', 'mutated_guide', 'strand', 'num_mismatch']) #CHECK FOR GRCh38 for i in range(variant_set.shape[0]): #ref_listC=listC[listC['gene']==variant_set.iloc[i]['gene']][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listC=ref_list[ref_list['sgID_AB']==variant_set.iloc[i]][['guide_type','protospacer_A','protospacer_B','sgID_AB']] ref_listC =ref_listC[['sgID_AB','guide_type','protospacer_A','protospacer_B']] #ref_listC.columns=['gene','guide_type','protospacer_A','protospacer_B'] res,res_mut,res_notfound,list_match,list_mutated,gflgc1=get_lists(ref_listC,list_found,list_notfound) #dft_c=dft_c.append(ref_listC) if res.shape[0]>0: dft_resc=pd.concat([dft_resc,res]) if res_mut.shape[0]>0: dft_res_mutc=pd.concat([dft_res_mutc,res_mut]) if res_notfound.shape[0]>0: dft_notfoundc= pd.concat([dft_notfoundc,res_notfound]) if list_match.shape[0]>0: df_matched_guides_ref= pd.concat([df_matched_guides_ref,list_match]) if list_mutated.shape[0]>0: df_mutated_guides_ref= pd.concat([df_mutated_guides_ref,list_mutated]) #st.write('Selected Reference Guides for **Set C**') #tbl_disp(dft_c,'All','ReferenceGuides',0) st.write('**Important:** If a guides is **not** in **found, mutated and not_found list (such as GSTT1), then it is found in Alternative Loci and Removed**') if dft_resc.shape[0]>0: st.write('Matched to '+ref_sel+' Reference Guides for **Set C**') tbl_disp(dft_resc,'select_genes','SetC_GRCh38',17) elif dft_res_mutc.shape[0]>0: st.write('None of the guides Matched, So reporting **Mutated to '+ref_sel+' Reference Guides for **Set C**') st.markdown(caution1,unsafe_allow_html=True) tbl_disp(dft_res_mutc,'select_genes','SetC_Mutated_GRCh38',18) if dft_notfoundc.shape[0]>0: st.write('**SetC Guides Not Found in '+ref_sel+' (None of the guides are Matched/Mutated)**') #tbl_disp(dft_notfound,'select_genes','SetA_Notfound_GRCh38') st.table(dft_notfoundc) elif Calc=='Not_Found': ListAResNotFound = st.checkbox('Results For SetA',key=30) if ListAResNotFound and listA_notfound_lr.shape[0]>0: listA_notfound_LR_sorted=listA_notfound_lr.sort_values('gene') sz1a=listA_notfound_LR_sorted.shape[0] vaild_guides_a = listA_notfound_LR_sorted[~listA_notfound_LR_sorted['gene'].str.contains("non")] sz2a=vaild_guides_a.shape[0] st.write(str(sz2a)+"/"+str(sz1a)+' Guides Not Found') tbl_disp(vaild_guides_a,'all_not_found','SetA_KOLF2.1',23,0) #now get gene names only genesa=vaild_guides_a['gene'].str.split('_').str[0] genesa1=genesa[genesa.duplicated(keep=False)] genesa2=genesa1.unique() pair_lista=[] for g in genesa2: g1=vaild_guides_a[vaild_guides_a['gene'].str.contains(g)] g2=g1.reset_index(drop=True) pair_lista.append([g2.gene[0],g2.ref_guide[0],g2.gene[1],g2.ref_guide[1]]) pair_missmatch_a = pd.DataFrame(pair_lista, columns=['sgID_1','sgRNA_1','sgID_2','sgRNA_2']) sz22a=pair_missmatch_a.shape[0] st.write(str(sz22a)+"/"+str(sz2a)+' Paired Guides Not Found') tbl_disp(pair_missmatch_a,'all_not_found','SetA_KOLF2.1',23,0) non_targeting_guides_a = listA_notfound_LR_sorted[listA_notfound_LR_sorted['gene'].str.contains("non")] sz3a=non_targeting_guides_a.shape[0] st.write(str(sz3a)+"/"+str(sz1a)+' no-targeting Guides Not Found') tbl_disp(non_targeting_guides_a,'all_not_found','SetA_KOLF2.1',23,0) ListBResNotFound = st.checkbox('Results For SetB',key=40) if ListBResNotFound: listB_notfound_LR_sorted=listB_notfound_lr.sort_values('gene') sz1b=listB_notfound_LR_sorted.shape[0] vaild_guides_b = listB_notfound_LR_sorted[~listB_notfound_LR_sorted['gene'].str.contains("non")] sz2b=vaild_guides_b.shape[0] st.write(str(sz2b)+"/"+str(sz1b)+' Guides Not Found') tbl_disp(vaild_guides_b,'all_not_found','SetA_KOLF2.1',23,0) #now get gene names only genesb=vaild_guides_b['gene'].str.split('_').str[0] genesb1=genesb[genesb.duplicated(keep=False)] genesb2=genesb1.unique() pair_listb=[] for g in genesb2: g1=vaild_guides_b[vaild_guides_b['gene'].str.contains(g)] g2=g1.reset_index(drop=True) pair_listb.append([g2.gene[0],g2.ref_guide[0],g2.gene[1],g2.ref_guide[1]]) pair_missmatch_b = pd.DataFrame(pair_listb, columns=['sgID_1','sgRNA_1','sgID_2','sgRNA_2']) sz22b=pair_missmatch_b.shape[0] st.write(str(sz22b)+"/"+str(sz2b)+' Paired Guides Not Found') tbl_disp(pair_missmatch_b,'all_not_found','SetA_KOLF2.1',23,0) non_targeting_guides_b = listB_notfound_LR_sorted[listB_notfound_LR_sorted['gene'].str.contains("non")] sz3b=non_targeting_guides_b.shape[0] st.write(str(sz3b)+"/"+str(sz1b)+' no-targeting Guides Not Found') tbl_disp(non_targeting_guides_b,'all_not_found','SetA_KOLF2.1',23,0) ListCResNotFound = st.checkbox('Results For SetC',key=50) if ListCResNotFound: listC_notfound_LR_sorted=listC_notfound_lr.sort_values('gene') sz1c=listC_notfound_LR_sorted.shape[0] vaild_guides_c = listC_notfound_LR_sorted[~listC_notfound_LR_sorted['gene'].str.contains("non")] sz2c=vaild_guides_c.shape[0] st.write(str(sz2c)+"/"+str(sz1c)+' Guides Not Found') tbl_disp(vaild_guides_c,'all_not_found','SetA_KOLF2.1',23,0) #now get gene names only genesc=vaild_guides_c['gene'].str.split('_').str[0] genesc1=genesc[genesc.duplicated(keep=False)] genesc2=genesc1.unique() pair_listc=[] for g in genesc2: g1=vaild_guides_c[vaild_guides_c['gene'].str.contains(g)] g2=g1.reset_index(drop=True) pair_listc.append([g2.gene[0],g2.ref_guide[0],g2.gene[1],g2.ref_guide[1]]) pair_missmatch_c = pd.DataFrame(pair_listc, columns=['sgID_1','sgRNA_1','sgID_2','sgRNA_2']) sz22c=pair_missmatch_c.shape[0] st.write(str(sz22c)+"/"+str(sz2c)+' Paired Guides Not Found') tbl_disp(pair_missmatch_c,'all_not_found','SetA_KOLF2.1',23,0) non_targeting_guides_c = listC_notfound_LR_sorted[listC_notfound_LR_sorted['gene'].str.contains("non")] sz3c=non_targeting_guides_c.shape[0] st.write(str(sz3c)+"/"+str(sz1c)+' no-targeting Guides Not Found') tbl_disp(non_targeting_guides_c,'all_not_found','SetA_KOLF2.1',23,0) else: guidetype = st.radio("Select Guide Type",('Non-targetting','Regular'),horizontal=True) if guidetype=='Non-targetting': with st.form(key='columns_in_form_non'): c2, c3 = st.columns([5,5])#([10,10]) with c2: guides_List = st.selectbox('Please select list', ('ListA','ListB','ListC')) with c3: ref_type_sel_non = st.radio("Select Reference", ('CHM13','GRCh38'), horizontal=True) Show_Results_non=st.form_submit_button(label = 'Non-targeting Guides Results') if Show_Results_non and guides_List=='ListA': for_list=listA if ref_type_sel_non=='GRCh38': f_list=listA_found_ref nf_list=listA_notfound_ref else: f_list=listA_found_lr nf_list=listA_notfound_lr st.write('Total: '+str(len(non_targeting_lista))+' Non-targeting Guide pairs and '+str(2*len(non_targeting_lista))+' single guides in ListA') process_all_guides(pd.DataFrame(pd.Series(non_targeting_lista,name='gene')),for_list,f_list,nf_list) if Show_Results_non and guides_List=='ListB': for_list=listB if ref_type_sel_non=='GRCh38': f_list=listB_found_ref nf_list=listB_notfound_ref else: f_list=listB_found_lr nf_list=listB_notfound_lr st.write('Total: '+str(len(non_targeting_listb))+' Non-targeting Guide pairs and '+str(2*len(non_targeting_listb))+' single guides in ListA') process_all_guides(pd.DataFrame(pd.Series(non_targeting_listb,name='gene')),for_list,f_list,nf_list) if Show_Results_non and guides_List=='ListC': for_list=listC if ref_type_sel_non=='GRCh38': f_list=listC_found_ref nf_list=listC_notfound_ref else: f_list=listC_found_lr nf_list=listC_notfound_lr st.write('Total: '+str(len(non_targeting_listc))+' Non-targeting Guide pairs and '+str(2*len(non_targeting_listc))+' single guides in ListA') process_all_guides(pd.DataFrame(pd.Series(non_targeting_listc,name='gene')),for_list,f_list,nf_list) elif guidetype=='Regular': st.write('**Maximum End Index=** '+str(regular_lista.shape[0])) with st.form(key='columns_in_form_regular'): c2, c3, c4 = st.columns([5,5,5])#([10,10]) with c2: set_start = int(st.text_input('Start Index', '0')) with c3: set_end = int(st.text_input('End Index', str(regular_lista.shape[0]))) with c4: ref_type_sel = st.radio("Select Reference", ('CHM13','GRCh38'), horizontal=True) Show_Results=st.form_submit_button(label = 'Show Regular Guides Results') if Show_Results:# and guides_List=="ListA": regular_listc=regular_listc[set_start:set_end] regular_listb=regular_listb.iloc[set_start:set_end] regular_lista=regular_lista.iloc[set_start:set_end] if ref_type_sel=='GRCh38': list_founda=listA_found_ref list_notfounda=listA_notfound_ref list_foundb=listB_found_ref list_notfoundb=listB_notfound_ref list_foundc=listC_found_ref list_notfoundc=listC_notfound_ref else: list_founda=listA_found_lr list_notfounda=listA_notfound_lr list_foundb=listB_found_lr list_notfoundb=listB_notfound_lr list_foundc=listC_found_lr list_notfoundc=listC_notfound_lr dupesq=list(duplicates(listA['gene'])) non_targetinga=variantsa1[pd.Series(variantsa1).str.contains('non-targeting')] regulara=variantsa1[~pd.Series(variantsa1).str.contains('non-targeting')] st.write('Total: '+str(len(regulara))+' Regular Guide (unique genes only) **Excluding:** '+str(len(non_targetinga))+' Non-targeting pairs **and** '+str(len(dupesq))+' Repeated entries (same gene names)') order_ready_tbl_CHM13(regular_lista,regular_listb,regular_listc,list_founda,list_notfounda,list_foundb,list_notfoundb,list_foundc,list_notfoundc,ref_type_sel)