gwaslab 3.5.3__py3-none-any.whl → 3.5.5__py3-none-any.whl

gwaslab/__init__.py

@@ -46,4 +46,5 @@ from gwaslab.util_ex_process_h5 import process_vcf_to_hfd5
 from gwaslab.util_ex_run_susie import _run_susie_rss as run_susie_rss
 from gwaslab.io_read_tabular import _read_tabular as read_tabular
 from gwaslab.util_in_meta import meta_analyze
-from gwaslab.viz_plot_scatter_with_reg import scatter
+from gwaslab.viz_plot_scatter_with_reg import scatter
+from gwaslab.util_in_fill_data import rank_based_int

gwaslab/g_Sumstats.py

@@ -80,6 +80,8 @@ from gwaslab.util_ex_ldsc import _estimate_partitioned_h2_by_ldsc
 from gwaslab.bd_get_hapmap3 import gethapmap3
 from gwaslab.util_abf_finemapping import abf_finemapping
 from gwaslab.util_abf_finemapping import make_cs
+from gwaslab.io_read_pipcs import _read_pipcs
+from gwaslab.viz_plot_credible_sets import _plot_cs
 import gc
 from gwaslab.viz_plot_phe_heatmap import _gwheatmap
 
@@ -88,6 +90,7 @@ class Sumstats():
     def __init__(self,
              sumstats,
              fmt=None,
+             tab_fmt="tsv",
              snpid=None,
              rsid=None,
              chrom=None,
@@ -154,10 +157,17 @@ class Sumstats():
         self.meta["gwaslab"]["species"] = species
         
         # initialize attributes for clumping and finmapping
-        self.to_finemapping_file_path = ""
-        self.to_finemapping_file  = pd.DataFrame()
-        self.plink_log = ""
-        self.clumps = pd.DataFrame()
+        #self.to_finemapping_file_path = ""
+        #self.to_finemapping_file  = pd.DataFrame()
+        #self.plink_log = ""
+
+        # path / file / plink_log
+        self.finemapping = dict()
+
+        # clumps / clumps_raw / plink_log
+        self.clumps = dict()
+        
+        #
         self.pipcs = pd.DataFrame()
 
         # print gwaslab version information
@@ -167,6 +177,7 @@ class Sumstats():
         self.data  = preformat(
           sumstats=sumstats,
           fmt=fmt,
+          tab_fmt = tab_fmt,
           snpid=snpid,
           rsid=rsid,
           chrom=chrom,
@@ -822,21 +833,31 @@ class Sumstats():
 # external ################################################################################################
     
     def calculate_ld_matrix(self,**kwargs):
-        self.to_finemapping_file_path, self.to_finemapping_file, self.plink_log  = tofinemapping(self.data,study = self.meta["gwaslab"]["study_name"],**kwargs)
+        self.finemapping["path"],self.finemapping["file"],self.finemapping["plink_log"]= tofinemapping(self.data,study = self.meta["gwaslab"]["study_name"],**kwargs)
+        #self.to_finemapping_file_path, self.to_finemapping_file, self.plink_log  = tofinemapping(self.data,study = self.meta["gwaslab"]["study_name"],**kwargs)
     
     def run_susie_rss(self,**kwargs):
-        self.pipcs=_run_susie_rss(self.to_finemapping_file_path,**kwargs)
+        self.pipcs=_run_susie_rss(self.finemapping["path"],**kwargs)
+        #self.pipcs=_run_susie_rss(self.to_finemapping_file_path,**kwargs)
     
     def clump(self,**kwargs):
-        self.clumps,self.plink_log = _clump(self.data, log=self.log, study = self.meta["gwaslab"]["study_name"], **kwargs)
+        self.clumps["clumps"], self.clumps["clumps_raw"], self.clumps["plink_log"] = _clump(self.data, log=self.log, study = self.meta["gwaslab"]["study_name"], **kwargs)
 
     def calculate_prs(self,**kwargs):
         combined_results_summary = _calculate_prs(self.data, log=self.log, study = self.meta["gwaslab"]["study_name"], **kwargs)
         return combined_results_summary
-    
+
+# loading aux data
+    def read_pipcs(self,prefix,**kwargs):
+        self.pipcs = _read_pipcs(self.data[["SNPID","CHR","POS"]],prefix, **kwargs)
+
+    def plot_pipcs(self, region,**kwargs):
+        _plot_cs(self.pipcs, region, **kwargs)
 # to_format ###############################################################################################       
 
     def to_format(self, path, build=None, verbose=True, **kwargs):
         if build is None:
             build = self.meta["gwaslab"]["genome_build"]
         _to_format(self.data, path, log=self.log, verbose=verbose, meta=self.meta, build=build, **kwargs)
+
+ 

gwaslab/g_SumstatsPair.py

@@ -23,6 +23,7 @@ from gwaslab.viz_plot_compare_af import  plotdaf
 from gwaslab.util_ex_run_2samplemr import _run_two_sample_mr
 from gwaslab.util_ex_run_clumping import _clump
 from gwaslab.util_ex_ldproxyfinder import _extract_with_ld_proxy
+from gwaslab.g_headers import _get_headers
 
 class SumstatsPair( ):
     def __init__(self, sumstatsObject1, sumstatsObject2, study=None, suffixes = ("_1","_2") ,verbose=True ):
@@ -35,6 +36,7 @@ class SumstatsPair( ):
             self.study_name = "{}_{}".format(sumstatsObject1.meta["gwaslab"]["study_name"], sumstatsObject2.meta["gwaslab"]["study_name"])
         else:
             self.study_name = "{}_{}".format("STUDY1", "STUDY2")
+        
         self.snp_info_cols = []
         self.stats_cols =[]
         self.stats_cols2 =[]
@@ -45,11 +47,13 @@ class SumstatsPair( ):
         self.colocalization=pd.DataFrame()
         self.sumstats1 = pd.DataFrame()
         self.sumstats2 = pd.DataFrame()
-        self.mr = {}
-        self.clumps ={}
+        
+        self.mr =dict()
+        self.clumps =dict()
         self.ns = None
-        self.to_finemapping_file_path = ""
-        self.plink_log = ""
+        self.finemapping = dict()
+        #self.to_finemapping_file_path = ""
+        #self.plink_log = ""
 
         self.log.write( "Start to create SumstatsPair object..." )
         
@@ -66,16 +70,17 @@ class SumstatsPair( ):
                                 verbose=verbose)
 
         for i in sumstatsObject1.data.columns:
-            if i in ["SNPID","rsID","CHR","POS","EA","NEA","STATUS"]:
+            if i in _get_headers(mode="info"):
+                # extract SNP info columns from sumstats1
                 self.snp_info_cols.append(i)
-            elif i in ["BETA","SE","P","MLOG10P","N","N_CASE","N_CONTROL","Z","T","F","OR","OR_95L","OR_95U","HR","HR_95L","HR_95U","MAF","EAF","BETA_95L","BETA_95U"]:
+            elif i in _get_headers(mode="stats"):
                 self.stats_cols.append(i)
             else:
                 self.other_cols.append(i)
         for i in sumstatsObject2.data.columns:
-            if i in ["SNPID","rsID","CHR","POS","EA","NEA","STATUS"]:
+            if i in _get_headers(mode="info"):
                 continue
-            elif i in ["BETA","SE","P","MLOG10P","N","N_CASE","N_CONTROL","Z","T","F","OR","OR_95L","OR_95U","HR","HR_95L","HR_95U","MAF","EAF","BETA_95L","BETA_95U"]:
+            elif i in _get_headers(mode="stats"):
                 self.stats_cols2.append(i)
             else:
                 self.other_cols2.append(i)           
@@ -136,14 +141,13 @@ class SumstatsPair( ):
 
 
     def clump(self,**kwargs):
-        self.clumps["clumps"], self.clumps["plink_log"] = _clump(self.data, log=self.log, p="P_1",mlog10p="MLOG10P_1", study = self.study_name, **kwargs)
+        self.clumps["clumps"],self.clumps["clumps_raw"],self.clumps["plink_log"] = _clump(self.data, log=self.log, p="P_1",mlog10p="MLOG10P_1", study = self.study_name, **kwargs)
 
     def to_coloc(self,**kwargs):
-        self.to_finemapping_file_path, output_file_list, self.plink_log = tofinemapping(self.data,study=self.study_name,suffixes=self.suffixes,log=self.log,**kwargs)
+        self.finemapping["path"],self.finemapping["file"],self.finemapping["plink_log"] = tofinemapping(self.data,study=self.study_name,suffixes=self.suffixes,log=self.log,**kwargs)
 
     def run_coloc_susie(self,**kwargs):
-
-        self.colocalization = _run_coloc_susie(self.to_finemapping_file_path,log=self.log,ncols=self.ns,**kwargs)
+        self.colocalization = _run_coloc_susie(self.finemapping["path"],log=self.log,ncols=self.ns,**kwargs)
 
     def run_two_sample_mr(self, clump=False, **kwargs):
         exposure1 = self.study_name.split("_")[0]

gwaslab/g_headers.py

@@ -0,0 +1,131 @@
+dtype_dic={
+ 'SNPID'             : 'string'   ,
+ 'rsID'              : 'string'   ,
+ 'CHR'               : 'Int64'    ,
+ 'POS'               : 'Int64'    ,
+ 'EA'                : 'category' ,
+ 'NEA'               : 'category' ,
+ 'STATUS'            : 'category' ,
+ 'REF'               : 'category' ,
+ 'ALT'               : 'category' ,
+ 'EAF'               : 'float64'  ,
+ 'NEAF'              : 'float64'  ,
+ 'MAF'               : 'float64'  ,
+ 'INFO'              : 'float32'  ,
+ 'BETA'              : 'float64'  ,
+ 'SE'                : 'float64'  ,
+ 'BETA_95U'          : 'float64'  ,
+ 'BETA_95L'          : 'float64'  ,
+ 'OR'                : 'float64'  ,
+ 'OR_95U'            : 'float64'  ,
+ 'OR_95L'            : 'float64'  ,
+ 'HR'                : 'float64'  ,
+ 'HR_95U'            : 'float64'  ,
+ 'HR_95L'            : 'float64'  ,
+ 'CHISQ'             : 'float64'  ,
+ 'Z'                 : 'float64'  ,
+ 'T'                 : 'float64'  ,
+ 'F'                 : 'float64'  ,
+ 'P'                 : 'float64'  ,
+ 'P_MANTISSA'        : 'float64'  ,
+ 'P_EXPONENT'        : 'float64'  ,
+ 'MLOG10P'           : 'float64'  ,
+ 'SNPR2'             : 'float64'  ,
+ 'DOF'               : 'Int64'    ,
+ 'P_HET'             : 'float64'  ,
+ 'I2_HET'            : 'float64'  ,
+ 'DENSITY'           : 'Int64'    ,
+ 'N'                 : 'Int64'    ,
+ 'N_CASE'            : 'Int64'    ,
+ 'N_CONTROL'         : 'Int64'    ,
+ 'GENENAME'          : 'string'   ,
+ 'CIS/TRANS'         : 'string'   ,
+ 'DISTANCE_TO_KNOWN' : 'Int64'    ,
+ 'LOCATION_OF_KNOWN' : 'string'   ,
+ 'KNOWN_ID'          : 'string'   ,
+ 'KNOWN_PUBMED_ID'   : 'string'   ,
+ 'KNOWN_AUTHOR'      : 'string'   ,
+ 'KNOWN_SET_VARIANT' : 'string'   ,
+ 'KNOWN_VARIANT'     : 'string'   ,
+ 'KNOWN_SET'         : 'string'   ,
+ 'NOVEL'             : 'string'   ,
+ 'PIP'               :' float64   ',
+ 'CREDIBLE_SET_INDEX': 'Int64'    ,
+ 'N_SNP'             : 'Int64'    ,
+ 'LOCUS'             : 'string'   ,
+ 'STUDY'             : 'string'   ,
+}
+
+
+description_dic={
+ 'SNPID'             :' variant ID (CHR:POS:NEA:EA)                    ',
+ 'rsID'              :' dbSNP rsID                                     ',
+ 'CHR'               :' chromosome number (X 23, Y 24, MT 25)          ',
+ 'POS'               :' base pair position                             ',
+ 'EA'                :' effect allele                                  ',
+ 'NEA'               :' non-effect allele                              ',
+ 'STATUS'            :' variant standardization & harmonization status ',
+ 'REF'               :' reference allele in reference genome           ',
+ 'ALT'               :' alternative allele                             ',
+ 'EAF'               :' effect allele frequency                        ',
+ 'NEAF'              :' non-effect allele frequency                    ',
+ 'MAF'               :' minor allele frequency                         ',
+ 'INFO'              :' imputation INFO/RSQ                            ',
+ 'BETA'              :' effect size beta                               ',
+ 'SE'                :' standard error of beta                         ',
+ 'BETA_95U'          :' upper bound of beta 95% condidence interval    ',
+ 'BETA_95L'          :' lower bound of beta 95% condidence interval    ',
+ 'OR'                :' odds ratio                                     ',
+ 'OR_95U'            :' upper bound of OR 95% condidence interval      ',
+ 'OR_95L'            :' lower bound of OR 95% condidence interval      ',
+ 'HR'                :' hazard ratio                                   ',
+ 'HR_95U'            :' upper bound of HR 95% condidence interval      ',
+ 'HR_95L'            :' lower bound of HR 95% condidence interval      ',
+ 'CHISQ'             :' chi square                                     ',
+ 'Z'                 :' z score                                        ',
+ 'T'                 :' t statistics                                   ',
+ 'F'                 :' F statistics                                   ',
+ 'P'                 :' P value                                        ',
+ 'P_MANTISSA'        :' P mantissa                                     ',
+ 'P_EXPONENT'        :' P exponent                                     ',
+ 'MLOG10P'           :' $-log_{10}(P)$                                 ',
+ 'SNPR2'             :' per variant R2                                 ',
+ 'DOF'               :' degree of freedom                              ',
+ 'P_HET'             :' heterogeneity test P value                     ',
+ 'I2_HET'            :' heterogeneity I2                               ',
+ 'DENSITY'           :' signal density                                 ',
+ 'N'                 :' total sample size                              ',
+ 'N_CASE'            :' number of cases                                ',
+ 'N_CONTROL'         :' number of controls                             ',
+ 'GENENAME'          :' nearest gene symbol                            ',
+ 'CIS/TRANS'         :' whether the variant is in cis or trans region  ',
+ 'DISTANCE_TO_KNOWN' :' distance to nearest known variants             ',
+ 'LOCATION_OF_KNOWN' :' relative location to nearest known variants    ',
+ 'KNOWN_ID'          :' nearest known variant ID                       ',
+ 'KNOWN_PUBMED_ID'   :' pubmed ID of the known variant                 ',
+ 'KNOWN_AUTHOR'      :' author of the study                            ',
+ 'KNOWN_SET_VARIANT' :' known set and overlapping variant              ',
+ 'KNOWN_VARIANT'     :' known variant overlapping with the variant     ',
+ 'KNOWN_SET'         :' variant set of the known variant               ',
+ 'PIP'               :' Posterior Inclusion Probability                ',
+ 'CREDIBLE_SET_INDEX':' credible sets index           ',
+ 'N_SNP'             :' number of variants included in this locus for finemapping           ',
+ 'LOCUS'             :' locus name, usually the lead variant of the locus           ',
+ 'STUDY'             :' study name           '}
+
+def _get_headers(mode="all"):
+    if mode=="info":
+        return ["SNPID","rsID","CHR","POS","EA","NEA","STATUS"]
+    elif mode=="stats":
+        return ["BETA","SE","P","MLOG10P","N","N_CASE","N_CONTROL","Z","T","F","OR","OR_95L","OR_95U","HR","HR_95L","HR_95U","MAF","EAF","BETA_95L","BETA_95U"]
+    else:
+        return description_dic.keys()
+
+def _check_overlap_with_reserved_keys(other):
+    overlapped=[]
+    for i in other:
+        if i in _get_headers():
+            overlapped.append(i)
+    return overlapped
+
+

gwaslab/g_meta.py

@@ -55,4 +55,5 @@ def _append_meta_record(old, new):
     if old == "Unknown" or old== "Unchecked":
         return new
     else:
-        return "{}, {}".format(old, new)
+        return "{}, {}".format(old, new)
+    

gwaslab/g_version.py

@@ -6,7 +6,7 @@ import numpy as np
 def _show_version(log=Log(), verbose=True):
     # show version when loading sumstats
     log.write("GWASLab v{} https://cloufield.github.io/gwaslab/".format(gwaslab_info()["version"]),verbose=verbose)
-    log.write("(C) 2022-2024, Yunye He, Kamatani Lab, MIT License, gwaslab@gmail.com",verbose=verbose)
+    log.write("(C) 2022-2025, Yunye He, Kamatani Lab, GPL-3.0 license, gwaslab@gmail.com",verbose=verbose)
 
 def _get_version():
     # return short version string like v3.4.33
@@ -15,8 +15,8 @@ def _get_version():
 def gwaslab_info():
     # version meta information
     dic={
-   "version":"3.5.3",
-   "release_date":"20241217"
+   "version":"3.5.5",
+   "release_date":"20250102"
     }
     return dic   
 

gwaslab/io_preformat_input.py

@@ -8,11 +8,13 @@ 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 import check_datatype
+from gwaslab.qc_check_datatype import quick_convert_datatype
 from gwaslab.qc_check_datatype import check_dataframe_memory_usage
-
+from gwaslab.g_headers import _check_overlap_with_reserved_keys
 #20221030
 def preformat(sumstats,
           fmt=None,
+          tab_fmt="tsv",
           snpid=None,
           rsid=None,
           chrom=None,
@@ -66,12 +68,21 @@ def preformat(sumstats,
     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
@@ -145,9 +156,11 @@ def preformat(sumstats,
             if key in raw_cols:
                 usecols.append(key)
             if value in ["EA","NEA"]:
-                dtype_dictionary[value]="category"
-            if value in ["CHR","STATUS"]:
-                dtype_dictionary[value]="string"     
+                dtype_dictionary[key]="category"
+            if value in ["STATUS"]:
+                dtype_dictionary[key]="string"     
+            if value in ["CHR"]:
+                dtype_dictionary[key]="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.")
@@ -276,6 +289,8 @@ def preformat(sumstats,
         rename_dictionary[status]="STATUS"
         dtype_dictionary[status]="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      
@@ -359,8 +374,13 @@ def preformat(sumstats,
             sumstats = sumstats[usecols].copy()
             for key,value in dtype_dictionary.items():
                 if key in usecols:
-                    sumstats[key] = sumstats[key].astype(value)
-
+                    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.")
 
@@ -400,6 +420,8 @@ def preformat(sumstats,
 
     ## 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)

gwaslab/io_read_pipcs.py

@@ -0,0 +1,23 @@
+import pandas as pd
+from gwaslab.g_Log import Log
+from gwaslab.qc_check_datatype import check_datatype
+from gwaslab.qc_check_datatype import check_dataframe_memory_usage
+
+def _read_pipcs(data, output_prefix, log=Log(),verbose=True):
+    log.write("Start to load PIP and CREDIBLE_SET_INDEX from file...",verbose=verbose)
+    log.write(" -File:{}.pipcs".format(output_prefix),verbose=verbose)
+
+    pipcs = pd.read_csv("{}.pipcs".format(output_prefix))
+    
+    log.write(" -Merging CHR and POS from main dataframe...",verbose=verbose)
+    pipcs = _merge_chrpos(data,pipcs)
+
+    log.write(" -Current pipcs Dataframe shape :",len(pipcs)," x ", len(pipcs.columns),verbose=verbose) 
+    check_datatype(pipcs,log=log,verbose=verbose)
+    check_dataframe_memory_usage(pipcs,log=log,verbose=verbose)
+    log.write("Finished loading PIP and CREDIBLE_SET_INDEX from file!",verbose=verbose)
+    return pipcs
+
+def _merge_chrpos(data,pipcs):
+    df = pd.merge(pipcs, data,on="SNPID",how="left")
+    return df

gwaslab/io_to_formats.py

@@ -114,48 +114,49 @@ def _to_format(sumstats,
             
     #######################################################################################################
     #formatting float statistics
-    onetime_log.write(" -Formatting statistics ...",verbose=verbose)
     
-    formats = {
-            'EAF': '{:.4g}', 
-            'MAF': '{:.4g}', 
-            'BETA': '{:.4f}', 
-            'SE': '{:.4f}',
-            'BETA_95U': '{:.4f}',
-            'BETA_95L': '{:.4f}',
-            'Z': '{:.4f}',
-            'CHISQ': '{:.4f}',
-            'F': '{:.4f}',
-            'OR': '{:.4f}',
-            'OR_95U': '{:.4f}',
-            'OR_95L': '{:.4f}',
-            'HR': '{:.4f}',
-            'HR_95U': '{:.4f}',
-            'HR_95L': '{:.4f}',
-            'INFO': '{:.4f}',
-            'P': '{:.4e}',
-            'MLOG10P': '{:.4f}',
-            'DAF': '{:.4f}'}
-    
-    for col, f in float_formats.items():
-        if col in output.columns: 
-            formats[col]=f
-    
-    for col, f in formats.items():
-        if col in output.columns: 
-            if str(output[col].dtype) in ["Float32","Float64","float64","float32","float16","float"]:
-                output[col] = output[col].map(f.format)
-    
-    onetime_log.write(" -Float statistics formats:",verbose=verbose)  
-    keys=[]
-    values=[]
-    for key,value in formats.items():
-        if key in output.columns: 
-            keys.append(key)
-            values.append(value)
-    
-    onetime_log.write("  - Columns       :",keys,verbose=verbose) 
-    onetime_log.write("  - Output formats:",values,verbose=verbose) 
+    if tab_fmt!="parquet":
+        onetime_log.write(" -Formatting statistics ...",verbose=verbose)
+        formats = {
+                'EAF': '{:.4g}', 
+                'MAF': '{:.4g}', 
+                'BETA': '{:.4f}', 
+                'SE': '{:.4f}',
+                'BETA_95U': '{:.4f}',
+                'BETA_95L': '{:.4f}',
+                'Z': '{:.4f}',
+                'CHISQ': '{:.4f}',
+                'F': '{:.4f}',
+                'OR': '{:.4f}',
+                'OR_95U': '{:.4f}',
+                'OR_95L': '{:.4f}',
+                'HR': '{:.4f}',
+                'HR_95U': '{:.4f}',
+                'HR_95L': '{:.4f}',
+                'INFO': '{:.4f}',
+                'P': '{:.4e}',
+                'MLOG10P': '{:.4f}',
+                'DAF': '{:.4f}'}
+        
+        for col, f in float_formats.items():
+            if col in output.columns: 
+                formats[col]=f
+        
+        for col, f in formats.items():
+            if col in output.columns: 
+                if str(output[col].dtype) in ["Float32","Float64","float64","float32","float16","float"]:
+                    output[col] = output[col].map(f.format)
+        
+        onetime_log.write(" -Float statistics formats:",verbose=verbose)  
+        keys=[]
+        values=[]
+        for key,value in formats.items():
+            if key in output.columns: 
+                keys.append(key)
+                values.append(value)
+        
+        onetime_log.write("  - Columns       :",keys,verbose=verbose) 
+        onetime_log.write("  - Output formats:",values,verbose=verbose) 
         
     ##########################################################################################################          
     # output, mapping column names
@@ -233,7 +234,7 @@ def tofmt(sumstats,
         if xymt_number is False and pd.api.types.is_integer_dtype(sumstats["CHR"]):
             sumstats["CHR"]= sumstats["CHR"].map(get_number_to_chr(xymt=xymt,prefix=chr_prefix))
         # add prefix to CHR
-        elif chr_prefix is not None:
+        elif len(chr_prefix)>0:
             sumstats["CHR"]= chr_prefix + sumstats["CHR"].astype("string")
 
     ####################################################################################################################
@@ -409,7 +410,7 @@ def _write_tabular(sumstats,rename_dictionary, path, tab_fmt, to_csvargs, to_tab
                 log.write(f"  -@ detected: writing each chromosome to a single file...",verbose=verbose)
                 log.write("  -Chromosomes:{}...".format(list(sumstats["CHR"].unique())),verbose=verbose)
                 for single_chr in list(sumstats["CHR"].unique()):
-                    single_path = path.replace("@",single_chr)
+                    single_path = path.replace("@","{}".format(single_chr))
 
                     fast_to_csv(sumstats.loc[sumstats[chr_header]==single_chr,:],
                                  single_path, 
@@ -422,7 +423,7 @@ def _write_tabular(sumstats,rename_dictionary, path, tab_fmt, to_csvargs, to_tab
                 log.write(f"  -@ detected: writing each chromosome to a single file...",verbose=verbose)
                 log.write("  -Chromosomes:{}...".format(list(sumstats["CHR"].unique())),verbose=verbose)
                 for single_chr in list(sumstats["CHR"].unique()):
-                    single_path = path.replace("@",single_chr)
+                    single_path = path.replace("@","{}".format(single_chr))
 
                     sumstats.loc[sumstats[chr_header]==single_chr,:].to_csv(path, index=None, **to_csvargs)
             else: