gwaslab 3.5.7__py3-none-any.whl → 3.5.8__py3-none-any.whl

gwaslab/io_preformat_input_polars.py

@@ -0,0 +1,632 @@
+import pandas as pd
+import polars as pl
+import numpy as np
+import scipy.stats as ss
+import gzip
+import os
+import gc
+from gwaslab.bd_common_data import get_format_dict
+from gwaslab.qc_fix_sumstats import sortcolumn
+from gwaslab.qc_fix_sumstats import _process_build
+from gwaslab.qc_check_datatype_polars import check_datatype
+from gwaslab.qc_check_datatype_polars import quick_convert_datatype
+from gwaslab.qc_check_datatype_polars import check_dataframe_memory_usage
+from gwaslab.g_headers import _check_overlap_with_reserved_keys
+#20221030
+def preformatp(sumstats,
+          fmt=None,
+          tab_fmt="tsv",
+          snpid=None,
+          rsid=None,
+          chrom=None,
+          pos=None,
+          ea=None,
+          nea=None,
+          ref=None,
+          alt=None,
+          eaf=None,
+          neaf=None,
+          maf=None,
+          n=None,
+          beta=None,
+          se=None,
+          chisq=None,
+          z=None,
+          f=None,
+          t=None,
+          p=None,
+          q=None,
+          mlog10p=None,
+          test=None,
+          info=None,
+          OR=None,
+          OR_95L=None,
+          OR_95U=None,
+          beta_95L=None,
+          beta_95U=None,
+          HR=None,
+          HR_95L=None,
+          HR_95U=None,
+          i2=None,
+          snpr2=None,
+          phet=None,
+          dof=None,
+          ncase=None,
+          ncontrol=None,
+          neff=None,
+          direction=None,
+          status=None,
+          study=None,
+          trait=None,
+          build=None,
+          other=[],
+          usekeys=None,
+          chrom_pat=None,
+          snpid_pat=None,
+          verbose=False,
+          readargs=None,
+          log=None):
+
+    #renaming dictionary
+    rename_dictionary = {}
+    usecols = []
+    dtype_dictionary ={}    
+    if readargs is None:
+        readargs={}
+ #######################################################################################################################################################
+    # workflow: 
+    # 1. formatbook
+    # 2. user specified header
+    # 3. usekeys
+    if tab_fmt=="parquet":
+        if type(sumstats) is str:
+            log.write("Start to load data from parquet file....",verbose=verbose)
+            log.write(" -path: {}".format(sumstats),verbose=verbose)
+            sumstats = pd.read_parquet(sumstats,**readargs) 
+            log.write("Finished loading parquet file into pd.DataFrame....",verbose=verbose)
+        else:
+            raise ValueError("Please input a path for parquet file.")
+    
+    if fmt is not None:
+        # loading format parameters
+        log.write("Start to load format from formatbook....",verbose=verbose)
+        
+        # load format data
+        meta_data,rename_dictionary = get_format_dict(fmt)
+        
+        ########## print format information################################################
+        print_format_info(fmt=fmt, meta_data=meta_data,rename_dictionary=rename_dictionary,verbose=verbose, log=log)
+        
+        if "format_separator" in meta_data.keys():
+            if "separator" not in readargs.keys():
+                readargs["separator"] = meta_data["format_separator"]
+            else:
+                if readargs["separator"] != meta_data["format_separator"]:
+                    log.write('  - format_separator will be changed to: "{}"'.format(readargs["separator"]),verbose=verbose)
+
+        if "format_na" in meta_data.keys():
+            readargs["null_values"] = meta_data["format_na"]
+        
+        if "format_comment" in meta_data.keys():
+            readargs["comment_prefix"] = meta_data["format_comment"]
+        
+        if "format_other_cols" in meta_data.keys():
+            other += meta_data["format_other_cols"]
+        
+        if "sep" not in readargs.keys():
+            readargs["separator"] = "\t"
+        
+#########################################################################################################################################################      
+    
+    # check chr-separated path / vcf / then print header.            
+    try:
+        if type(sumstats) is str:
+            ## loading data from path #################################################
+            inpath = sumstats
+            ###load sumstats by each chromosome #################################################  
+            if "@" in inpath:
+                log.write(" -Detected @ in path: load sumstats by each chromosome...",verbose=verbose)
+                inpath_chr_list=[]
+                inpath_chr_num_list=[]
+                for chromosome in list(range(1,26))+["x","y","X","Y","MT","mt","m","M"]:
+                    inpath_chr = inpath.replace("@",str(chromosome))  
+                    if isfile_casesensitive(inpath_chr):
+                        inpath_chr_num_list.append(str(chromosome))
+                        inpath_chr_list.append(inpath_chr)        
+                log.write(" -Chromosomes detected:",",".join(inpath_chr_num_list),verbose=verbose)
+                readargs_header = get_readargs_header(inpath = inpath_chr_list[0], readargs = readargs)
+                row_one = pl.read_csv(inpath_chr_list[0],**readargs_header)
+                # columns in the sumstats
+                raw_cols = row_one.columns
+            else:
+            ##### loading data from tabular file#################################################
+                readargs_header = get_readargs_header(inpath = inpath, readargs = readargs)
+                row_one = pl.read_csv(inpath,**readargs_header)
+                raw_cols = row_one.columns
+            
+            if fmt=="vcf":
+                # expanded
+                format_cols = list(row_one["FORMAT"].str.split(":"))[0]
+                # fixed + study1 + expanded
+                raw_cols = meta_data["format_fixed"] + [raw_cols[9]] + format_cols
+
+            ###################################################################################### 
+        elif type(sumstats) is pd.DataFrame:
+            ## loading data from dataframe
+            raw_cols = sumstats.columns
+        
+        ################################################
+        for key,value in rename_dictionary.items():
+            # check avaiable keys  key->raw header
+            # usecols : a list of raw headers to load from file/DataFrame 
+            if key in raw_cols:
+                usecols.append(key)
+            if value in ["EA","NEA"]:
+                dtype_dictionary[key]=pl.String()
+            if value in ["STATUS"]:
+                dtype_dictionary[key]=pl.String()
+            if value in ["CHR"]:
+                dtype_dictionary[key]=pl.String()
+    
+    except ValueError:
+        raise ValueError("Please input a path or a pd.DataFrame, and make sure the separator is correct and the columns you specified are in the file.")
+        
+    ###################################################################################################################################################
+    ## check columns/datatype to use 
+    if snpid:
+        usecols.append(snpid)
+        rename_dictionary[snpid]= "SNPID"
+    if rsid:
+        usecols.append(rsid)
+        rename_dictionary[rsid]= "rsID"
+    if chrom:
+        usecols.append(chrom)
+        rename_dictionary[chrom]= "CHR"
+        dtype_dictionary[chrom]=pl.String()
+    if pos:
+        usecols.append(pos)
+        rename_dictionary[pos]= "POS"
+        dtype_dictionary[pos]=pl.Float64()
+    if ea:
+        usecols.append(ea)
+        rename_dictionary[ea]= "EA"
+        dtype_dictionary[ea]=pl.String()
+    if nea:
+        usecols.append(nea)
+        rename_dictionary[nea]= "NEA"
+        dtype_dictionary[nea]=pl.String()
+    if ref:
+        usecols.append(ref)
+        rename_dictionary[ref]= "REF"
+        dtype_dictionary[ref]=pl.String()
+    if alt:
+        usecols.append(alt)
+        rename_dictionary[alt]= "ALT"
+        dtype_dictionary[alt]=pl.String()
+    if eaf:
+        usecols.append(eaf)
+        rename_dictionary[eaf]= "EAF"
+    elif neaf:
+        # neaf will be converted to eaf
+        usecols.append(neaf)
+        rename_dictionary[neaf]= "EAF"
+    if maf:
+        usecols.append(maf)
+        rename_dictionary[maf]= "MAF"
+    if n and (type(n) is str):
+        usecols.append(n)
+        rename_dictionary[n]= "N"
+    if ncase and (type(ncase) is str):
+        usecols.append(ncase)
+        rename_dictionary[ncase]= "N_CASE"    
+    if ncontrol and (type(ncontrol) is str):
+        usecols.append(ncontrol)
+        rename_dictionary[ncontrol]= "N_CONTROL"    
+    if neff and (type(neff) is str):
+        usecols.append(neff)
+        rename_dictionary[neff]= "N_EFF" 
+    if beta:
+        usecols.append(beta)
+        rename_dictionary[beta]= "BETA"
+    if beta_95L:
+        usecols.append(beta_95L)
+        rename_dictionary[beta_95L]= "BETA_95L"
+    if beta_95U:
+        usecols.append(beta_95U)
+        rename_dictionary[beta_95U]= "BETA_95U"  
+    if se:
+        usecols.append(se)
+        rename_dictionary[se]= "SE"
+    if chisq:
+        usecols.append(chisq)
+        rename_dictionary[chisq]="CHISQ"
+    if z:
+        usecols.append(z)
+        rename_dictionary[z]= "Z"
+    if q:
+        usecols.append(q)
+        rename_dictionary[q]= "Q"
+    if p:
+        usecols.append(p)
+        rename_dictionary[p]= "P"
+    if t:
+        usecols.append(t)
+        rename_dictionary[t]= "T"
+    if f:
+        usecols.append(f)
+        rename_dictionary[f]= "F"
+    if mlog10p:
+        usecols.append(mlog10p)
+        rename_dictionary[mlog10p]= "MLOG10P"
+    if test:
+        usecols.append(test)
+        rename_dictionary[test]= "TEST"        
+    if info:
+        usecols.append(info)
+        rename_dictionary[info]= "INFO"
+    if OR:
+        usecols.append(OR)
+        rename_dictionary[OR]= "OR"
+    if OR_95L:
+        usecols.append(OR_95L)
+        rename_dictionary[OR_95L]= "OR_95L"
+    if OR_95U:
+        usecols.append(OR_95U)
+        rename_dictionary[OR_95U]= "OR_95U"
+    if HR:
+        usecols.append(HR)
+        rename_dictionary[HR]= "HR"
+    if HR_95L:
+        usecols.append(HR_95L)
+        rename_dictionary[HR_95L]= "HR_95L"
+    if HR_95U:
+        usecols.append(HR_95U)
+        rename_dictionary[HR_95U]= "HR_95U"        
+    if phet:
+        usecols.append(phet)
+        rename_dictionary[phet]= "P_HET"      
+    if i2:
+        usecols.append(i2)
+        rename_dictionary[i2]= "I2"    
+    if snpr2:
+        usecols.append(snpr2)
+        rename_dictionary[snpr2]= "SNPR2"    
+    if dof:
+        usecols.append(dof)
+        rename_dictionary[dof]= "DOF"    
+    if direction:
+        usecols.append(direction)
+        rename_dictionary[direction]="DIRECTION"
+    if status:
+        usecols.append(status)
+        rename_dictionary[status]="STATUS"
+        dtype_dictionary[status]=pl.String()
+    if other:
+        overlapped = _check_overlap_with_reserved_keys(other)
+        log.warning("Columns with headers overlapping with GWASLab reserved keywords:{}".format(overlapped),verbose=verbose)
+        usecols = usecols + other
+        for i in other:
+            rename_dictionary[i] = i      
+    if fmt=="vcf":
+        # store the final column list
+        vcf_usecols = usecols.copy()
+        # loading the fixed columns + study
+        usecols = meta_data["format_fixed"]
+        if study is not None:
+            usecols =  usecols + [study]
+        else:
+            study = raw_cols[9]
+            usecols =  usecols + [study]
+
+    if usekeys is not None:
+    # extract only specified keys
+        usecols_new =[]
+        for i in usekeys:
+            for k, v in rename_dictionary.items():
+                if i == v:
+                    usecols_new.append(k)
+        usecols_valid =[]
+        for i in usecols_new:
+            if i in usecols:
+                usecols_valid.append(i)
+        usecols = usecols_valid
+
+    usecols = list(set(usecols))
+
+ #loading data ##########################################################################################################
+    
+    try:
+        if type(sumstats) is str:
+            ## loading data from path
+            inpath = sumstats
+            if "@" in inpath:
+                log.write("Start to initialize gl.Sumstats from files with pattern :" + inpath,verbose=verbose)
+                sumstats_chr_list=[]
+                for i in inpath_chr_list:
+                    log.write(" -Loading:" + i)
+                    skip_rows = get_skip_rows(i)
+                    readargs["skip_rows"] = skip_rows
+                    sumstats_chr = pl.read_csv(i,
+                                        columns = usecols,
+                                        schema_overrides=dtype_dictionary,
+                                        **readargs)
+                    sumstats_chr_list.append(sumstats_chr)
+                log.write(" -Merging sumstats for chromosomes:",",".join(inpath_chr_num_list),verbose=verbose)
+                sumstats = pl.concat(sumstats_chr_list, rechunk=True) 
+                del(sumstats_chr_list)
+                gc.collect()
+            else:
+                skip_rows = get_skip_rows(inpath)
+                readargs["skip_rows"] = skip_rows
+                log.write("Start to initialize gl.Sumstats from file :" + inpath,verbose=verbose)
+
+                sumstats = pl.read_csv(inpath,
+                                    columns =usecols,
+                                    schema_overrides=dtype_dictionary,
+                                    **readargs)
+                
+        elif type(sumstats) is pd.DataFrame:
+            ## loading data from dataframe
+            log.write("Start to initialize gl.Sumstats from pandas DataFrame ...",verbose=verbose)
+            sumstats = sumstats[usecols].copy()
+            for key,value in dtype_dictionary.items():
+                if key in usecols:
+                    astype = value
+                    if rename_dictionary[key]=="CHR":
+                        astype ="Int64"  
+                    try:
+                        sumstats[key] = sumstats[key].astype(astype)
+                    except:
+                        sumstats[key] = sumstats[key].astype("string")
+    except ValueError:
+        raise ValueError("Please input a path or a pd.DataFrame, and make sure it contain the columns.")
+    
+    if chrom_pat is not None:
+        sumstats = _load_single_chr(sumstats,
+                                    usecols=usecols,
+                                    rename_dictionary=rename_dictionary,
+                                    chrom_pat=chrom_pat,
+                                    log=log,
+                                    verbose=verbose)
+    elif snpid_pat is not None: 
+        sumstats = _load_variants_with_pattern(sumstats,
+                                    usecols=usecols,
+                                    rename_dictionary=rename_dictionary,
+                                    snpid_pat=snpid_pat,
+                                    log=log,
+                                    verbose=verbose)
+    ## renaming columns ###############################################################################################
+    if fmt == "vcf":
+        sumstats = parse_vcf_study(sumstats,format_cols,study,vcf_usecols,log=log,verbose=verbose)
+        usecols = vcf_usecols
+    
+    converted_columns = list(map(lambda x: rename_dictionary[x], set(usecols)))
+    
+    ## renaming log
+    log.write(" -Reading columns          :", ",".join(set(usecols)),verbose=verbose)
+    log.write(" -Renaming columns to      :", ",".join(converted_columns),verbose=verbose)
+    log.write(" -Current Dataframe shape :",len(sumstats)," x ", len(sumstats.columns),verbose=verbose) 
+    
+    ## renaming  #####################################################################################
+    sumstats = sumstats.rename(rename_dictionary)
+
+    ## if n was provided as int #####################################################################################
+    if type(n) is int:
+        sumstats["N"] = n
+    if type(ncase) is int:
+        sumstats["N_CASE"] = ncase
+    if type(ncontrol) is int:
+        sumstats["N_CONTROL"] = ncontrol
+    
+    ### status ######################################################################################################
+    if status is None:
+        sumstats = process_status(sumstats=sumstats,build=build,log=log,verbose=verbose)
+    
+    ## ea/nea, ref/alt ##############################################################################################
+    sumstats = process_allele(sumstats=sumstats,log=log,verbose=verbose)
+        
+    ## NEAF to EAF ###########################################################################################################
+    if neaf is not None :
+        sumstats = process_neaf(sumstats=sumstats,log=log,verbose=verbose)
+
+    ## reodering ###################################################################################################  
+    #sumstats = sortcolumn(sumstats=sumstats,log=log,verbose=verbose)    
+    sumstats = quick_convert_datatype(sumstats,log=log,verbose=verbose)
+    
+    check_datatype(sumstats,log=log,verbose=verbose)
+    #gc.collect()
+    check_dataframe_memory_usage(sumstats,log=log,verbose=verbose)
+    
+    log.write("Finished loading data successfully!",verbose=verbose)
+    return sumstats
+
+
+#### helper #######################################################################
+def isfile_casesensitive(path):
+    if not os.path.isfile(path): 
+        return False   # exit early
+    directory, filename = os.path.split(path)
+    return filename in os.listdir(directory)
+
+def get_readargs_header(inpath,readargs):
+    if "vcf.gz" in inpath:
+        with gzip.open(inpath,'r') as file:      
+            skip=0
+            for line in file:        
+                if line.decode('utf-8').startswith('##'):
+                    skip+=1
+                else:
+                    readargs["skip_rows"]=skip
+                    readargs["separator"]="\t"
+                    break
+    readargs_header = readargs.copy()
+    readargs_header["n_rows"]=1
+    #readargs_header["dtype"]="string"
+    readargs_header["infer_schema"] = False
+    return readargs_header
+
+def get_skip_rows(inpath):
+    if "vcf.gz" in inpath:
+        with gzip.open(inpath,'r') as file:      
+            skip=0
+            for line in file:        
+                if line.decode('utf-8').startswith('##'):
+                    skip+=1
+                else:
+                    return skip
+    else:
+        return 0
+
+def parse_vcf_study(sumstats,format_cols,study,vcf_usecols,log,verbose=True):
+    log.write(" -Parsing based on FORMAT: ", format_cols,verbose=verbose)
+    log.write(" -Parsing vcf study : ", study,verbose=verbose)
+    #sumstats[format_cols] = sumstats[study].str.split(":",expand=True).values
+    sumstats = sumstats.drop(["FORMAT",study])
+    sumstats = sumstats[vcf_usecols]
+    gc.collect()
+    return sumstats
+
+def print_format_info(fmt,meta_data, rename_dictionary, verbose, log,output=False, skip_meta_records=None):
+    log.write(" -"+fmt+" format meta info:",verbose=verbose)   
+    if skip_meta_records is None:
+        skip_meta_records =[]
+    for key,value in meta_data.items():
+        if key in skip_meta_records:
+            continue
+        if value is None:
+            continue
+        if type(value) is str:
+            if "\n" in value:
+                value_first_line=value.split("\n")[0]
+                log.write("  -",key," : "+value_first_line.strip()+"...",verbose=verbose)
+            elif value==" ":
+                log.write('  -',key,' : \\s ',verbose=verbose)     
+            elif value=="\t":
+                log.write('  -',key,' : \\t',verbose=verbose)    
+            else:
+                log.write("  -",key," : "+value.strip(),verbose=verbose)  
+        elif type(value) is list:
+            log.write("  -",key," : "+','.join(value),verbose=verbose)  
+        else:
+            log.write("  -",key," : ",value,verbose=verbose)  
+    keys=[]
+    values=[]
+    for key,value in rename_dictionary.items():
+        keys.append(key)
+        values.append(value)
+    if fmt!="gwaslab":
+        if output == False:
+            if fmt!="auto":
+                log.write(" -"+fmt+" to gwaslab format dictionary:",verbose=verbose)  
+                log.write("  - "+fmt+" keys:",",".join(keys),verbose=verbose) 
+                log.write("  - gwaslab values:",",".join(values),verbose=verbose) 
+            else:
+                log.write("  - Auto-detection mode. Note: auto-detection assumes A1=EA; Alt=EA and Frq=EAF...",verbose=verbose)
+                log.write("  - Header conversion source: https://github.com/Cloufield/formatbook/blob/main/formats/auto.json",verbose=verbose)  
+        else:
+            log.write(" -gwaslab to "+fmt+" format dictionary:",verbose=verbose)  
+            keys=[]
+            values=[]
+            for key,value in rename_dictionary.items():
+                keys.append(key)
+                values.append(value)
+            log.write("  - gwaslab keys:",  ','.join(keys),verbose=verbose) 
+            log.write("  - "+fmt+" values:"  , ','.join(values),verbose=verbose)       
+
+def process_neaf(sumstats,log,verbose):
+    log.write(" -NEAF is specified...",verbose=verbose) 
+    pre_number=len(sumstats)
+    log.write(" -Checking if 0<= NEAF <=1 ...",verbose=verbose) 
+
+    sumstats["EAF"] = pd.to_numeric(sumstats["EAF"], errors='coerce')
+    
+    sumstats = sumstats.filter(pl.col("EAF")>=0 & pl.col("EAF")<=1)
+    sumstats = sumstats.with_columns(
+        EAF = 1- pl.col("EAF")
+    )
+    log.write(" -Converted NEAF to EAF.",verbose=verbose) 
+
+    after_number=len(sumstats)
+
+    log.write(" -Removed "+str(pre_number - after_number)+" variants with bad NEAF.",verbose=verbose) 
+    
+    return sumstats
+
+def process_allele(sumstats,log,verbose):
+    
+    if "EA" in sumstats.columns:
+
+        if "REF" in sumstats.columns and "ALT" in sumstats.columns:
+
+            if "NEA" not in sumstats.columns:
+                log.write(" NEA not available: assigning REF to NEA...",verbose=verbose) 
+
+                sumstats = sumstats.with_columns(NEA = pl.col("REF"))
+            
+            log.write(" -EA,REF and ALT columns are available: assigning NEA...",verbose=verbose) 
+            ea_alt = sumstats["EA"]==sumstats["ALT"]
+            
+            log.write(" -For variants with EA == ALT : assigning REF to NEA ...",verbose=verbose) 
+            sumstats.loc[ea_alt,"NEA"] = sumstats.loc[ea_alt,"REF"]
+            
+            sumstats = sumstats.with_columns(
+                pl.when(ea_alt)
+                .then(pl.col("REF"))
+                .otherwise(pl.col("NEA"))
+                .alias("NEA")
+                )
+
+            ea_not_alt = sumstats["EA"]!=sumstats["ALT"]
+            log.write(" -For variants with EA != ALT : assigning ALT to NEA ...",verbose=verbose) 
+            sumstats = sumstats.with_columns(
+                pl.when(ea_not_alt)
+                .then(pl.col("ALT"))
+                .otherwise(pl.col("NEA"))
+                .alias("NEA")
+                )
+            
+    return sumstats
+
+def process_status(sumstats,build,log,verbose):
+    log.write(" -Initiating a status column: STATUS ...",verbose=verbose)
+    #sumstats["STATUS"] = int(build)*(10**5) +99999
+    build = _process_build(build,log,verbose)
+    sumstats = sumstats.with_columns(
+        STATUS = pl.lit(build +"99999")
+    )
+    return sumstats
+
+
+def _load_single_chr(sumstats,usecols, rename_dictionary,chrom_pat,log,verbose):
+    
+
+    # get chr 
+    for k,v in rename_dictionary.items():
+        if v=="CHR":
+            if k in sumstats.columns:
+                log.write(" -Columns used to filter variants: {}".format(k),verbose=verbose)
+                chunk_chrom = k
+                break
+
+    log.write(" -Loading only variants on chromosome with pattern : {} ...".format(chrom_pat),verbose=verbose)
+    
+    sumstats_filtered = sumstats.filter(pl.col(chunk_chrom).str.contains(chrom_pat))
+
+    log.write(" -Loaded {} variants on chromosome with pattern :{} ...".format(len(sumstats_filtered), chrom_pat),verbose=verbose)
+    return sumstats_filtered
+
+def _load_variants_with_pattern(sumstats,usecols, rename_dictionary,snpid_pat,log,verbose):
+    
+    # get chr 
+    for k,v in rename_dictionary.items():
+        if v=="SNPID":
+            if k in sumstats.columns:
+                log.write(" -Columns used to filter variants: {}".format(k),verbose=verbose)
+                chunk_snpid = k
+                break
+
+    log.write(" -Loading only variants with pattern :  {} ...".format(snpid_pat),verbose=verbose)
+    sumstats_filtered = sumstats.filter(pl.col(chunk_snpid).str.contains(snpid_pat))
+    
+    log.write(" -Loaded {} variants with pattern : {} ...".format(len(sumstats_filtered), snpid_pat),verbose=verbose)
+    return sumstats_filtered

gwaslab/io_process_args.py

@@ -37,4 +37,28 @@ def _merge_and_sync_dic(list_of_dics:list, default:dict) -> dict:
     for dic in list_of_dics:
         if isinstance(dic, dict):
             temp.update(dic)
-    return temp
+    return temp
+
+def _update_args(args=None, default_args=None):
+    
+    if default_args is None:
+        default_args={}
+
+    if args is None:
+        # if None, return default dict
+        return default_args
+    else:
+        # if not None, update default dict
+        for key,value in args.items():
+            default_args[key] = value
+        return default_args
+
+    
+
+def _update_arg(arg=None, default_arg=None):
+    if arg is None:
+        # if None, return default
+        return default_arg
+    else:
+        # if not None, return arg  
+        return arg