gwaslab 3.4.38__py3-none-any.whl → 3.4.39__py3-none-any.whl

gwaslab/util_ex_calculate_ldmatrix.py

@@ -199,7 +199,7 @@ def _align_sumstats_with_bim(row, locus_sumstats, ref_bim, log=Log(),suffixes=No
     log.write("   -#Total Variants matched:{}".format(sum(allele_match)))
 
     if row["SNPID"] not in combined_df.loc[perfect_match,"SNPID"].values:
-        log.write("   -Warning: Lead variant was not available in reference!!!!!!!!!!!!!!!")
+        log.warning("Lead variant was not available in reference!")
     
     # adjust statistics
     output_columns=["SNPID","CHR","POS","EA_bim","NEA_bim"]
@@ -256,4 +256,4 @@ def _check_snpid_order(snplist_path, matched_sumstats_snpid,log):
     if list(matched_sumstats_snpid) == list(snpid_list):
         log.write(" -Sumstats SNPID order and LD matrix SNPID order are matched.")
     else:
-        log.write(" -Warning: Sumstats SNPID order and LD matrix SNPID order are not matched...")
+        log.warning("Sumstats SNPID order and LD matrix SNPID order are not matched!")

gwaslab/util_ex_gwascatalog.py

@@ -9,26 +9,26 @@ def gwascatalog_trait(efo,source="NCBI",sig_level=5e-8,verbose=True,log=Log()):
     #https://www.ebi.ac.uk/gwas/rest/docs/api
     
     base_url = "https://www.ebi.ac.uk/gwas/rest/api/efoTraits/"+efo
-    if verbose: log.write("Start to retrieve data from GWASCatalog...")
-    if verbose: log.write(" -Please make sure your sumstats is based on GRCh38...")
-    if verbose: log.write(" -Requesting (GET) trait information through the GWASCatalog API...")
-    if verbose: log.write(" -EFO trait api: "+ base_url)
+    log.write("Start to retrieve data from GWASCatalog...", verbose=verbose)
+    log.write(" -Please make sure your sumstats is based on GRCh38...", verbose=verbose)
+    log.write(" -Requesting (GET) trait information through the GWASCatalog API...", verbose=verbose)
+    log.write(" -EFO trait api: "+ base_url, verbose=verbose)
     text = requests.get(base_url)
-    if verbose: 
-        log.write(" -Status code: {}".format(text.status_code)) 
-        if text.status_code!=200:
-            log.write(" -Status code is not 200. Access failed. Please check your internet or the GWAS Catalog sever status.") 
-            log.write(" -Message:{}".format(text.text)) 
-            return 0
+
+    log.write(" -Status code: {}".format(text.status_code), verbose=verbose) 
+    if text.status_code!=200:
+        log.write(" -Status code is not 200. Access failed. Please check your internet or the GWAS Catalog sever status.", verbose=verbose) 
+        log.write(" -Message:{}".format(text.text), verbose=verbose) 
+        return 0
 
     api_response = json.loads(text.text)
-    if verbose: log.write(" -Trait Name:",api_response["trait"])
-    if verbose: log.write(" -Trait URL:",api_response["uri"]) 
+    log.write(" -Trait Name:",api_response["trait"], verbose=verbose)
+    log.write(" -Trait URL:",api_response["uri"], verbose=verbose) 
         
     base_url = "https://www.ebi.ac.uk/gwas/rest/api/efoTraits/"+efo+"/associations?projection=associationByEfoTrait"    
-    if verbose: log.write(" -Requesting (GET) GWAS associations through the GWASCatalog API...")
-    if verbose: log.write(" -associationsByTraitSummary API: "+ base_url)   
-    if verbose: log.write(" -Note: this step might take a while...")   
+    log.write(" -Requesting (GET) GWAS associations through the GWASCatalog API...", verbose=verbose)
+    log.write(" -associationsByTraitSummary API: "+ base_url, verbose=verbose)   
+    log.write(" -Note: this step might take a while...", verbose=verbose)   
     
     # get request and check status code of response
     raw_data = requests.get(base_url)
@@ -37,13 +37,13 @@ def gwascatalog_trait(efo,source="NCBI",sig_level=5e-8,verbose=True,log=Log()):
     is_proceed = check_request_status_code(raw_data.status_code,verbose=verbose,log=log)
     if is_proceed is False: return False
     
-    if verbose: log.write(" -Loading json ...")
+    log.write(" -Loading json ...", verbose=verbose)
     # Transform API response from JSON into Python dictionary
     api_response = json.loads(raw_data.text)
-    if verbose: log.write(" -Parsing json ...")        
+    log.write(" -Parsing json ...", verbose=verbose)        
     # An 
     records=list()
-    if verbose: log.write(" -Number of reported associations for "+ efo +" in GWASCatalog:",len( api_response["_embedded"]["associations"]))
+    log.write(" -Number of reported associations for "+ efo +" in GWASCatalog:",len( api_response["_embedded"]["associations"]), verbose=verbose)
    
     for association in api_response["_embedded"]["associations"]:
         #association statistics:       
@@ -126,12 +126,12 @@ def gwascatalog_trait(efo,source="NCBI",sig_level=5e-8,verbose=True,log=Log()):
                                 records.append(row)
             #rsid locations
     gwascatalog_lead_snps = pd.DataFrame(records,columns=["SNPID","CHR","POS","REPORT_GENENAME","CLOSEST_GENENAMES","FUNCTION_CLASS","OR","BETA","SE","P","TRAIT","STUDY","PUBMEDID","AUTHOR"])
-    if verbose: log.write(" -Loading retrieved data into gwaslab Sumstats object ...")  
+    log.write(" -Loading retrieved data into gwaslab Sumstats object ...", verbose=verbose)  
     sigs = gl.Sumstats(gwascatalog_lead_snps.copy(),fmt="gwaslab",other=['REPORT_GENENAME', 'CLOSEST_GENENAMES','TRAIT', 'STUDY', 'PUBMEDID','AUTHOR'],verbose=False)
     sigs.fix_pos(verbose=False)
     sigs.fix_chr(verbose=False)
     sigs.sort_coordinate(verbose=False)
-    if verbose: log.write("Finished retrieving data from GWASCatalog...")
+    log.write("Finished retrieving data from GWASCatalog...", verbose=verbose)
     #return gwaslab Sumstats object
     return sigs
 
@@ -142,14 +142,14 @@ def check_request_status_code(request_code,verbose=True,log=Log()):
     is_proceed=False
     
     if request_code == 200:
-        if verbose: log.write(" -Status code 200 OK: Retrieved data from GWASCatalog successffully ...")
+        log.write(" -Status code 200 OK: Retrieved data from GWASCatalog successffully ...", verbose=verbose)
         is_proceed=True
     elif request_code == 404:
-        if verbose: log.write(" -Status code 404 Not Found: The requested resource did not exist ...")
+        log.write(" -Status code 404 Not Found: The requested resource did not exist ...", verbose=verbose)
     elif request_code == 301:
-        if verbose: log.write(" -Status code 301 Moved Permanently: The requested resource did not exist ...")
+        log.write(" -Status code 301 Moved Permanently: The requested resource did not exist ...", verbose=verbose)
     elif request_code == 400:
-        if verbose: log.write(" -Status code 400 Bad Request: The requested resource did not exist ...")
+        log.write(" -Status code 400 Bad Request: The requested resource did not exist ...", verbose=verbose)
     
     return is_proceed
 

gwaslab/util_ex_ldproxyfinder.py

@@ -49,8 +49,8 @@ def _extract_with_ld_proxy(  snplist=None,
                             ld_threshold=0.8
                             ):
     ### Load vcf#######################################################################################
-    if verbose: log.write("Start to load reference genotype...")
-    if verbose: log.write(" -reference vcf path : "+ vcf_path)
+    log.write("Start to load reference genotype...", verbose=verbose)
+    log.write(" -reference vcf path : "+ vcf_path, verbose=verbose)
     if tabix is None:
         tabix = which("tabix")
     vcf_chr_dict = auto_check_vcf_chr_dict(vcf_path=vcf_path, vcf_chr_dict=vcf_chr_dict, verbose=verbose, log=log)
@@ -122,7 +122,7 @@ def _extract_with_ld_proxy(  snplist=None,
     
     extracted_sumstats = pd.concat([extracted_sumstats, ld_proxies],ignore_index=True)
 
-    if verbose: log.write("Finished loading reference genotype successfully!")
+    log.write("Finished loading reference genotype successfully!", verbose=verbose)
     return extracted_sumstats
 
 
@@ -139,13 +139,13 @@ def _get_rsq(    row,
         ref_genotype = read_vcf(vcf_path,region=vcf_chr_dict[region[0]]+":"+str(region[1])+"-"+str(region[2]),tabix=tabix)
         
         if ref_genotype is None:
-            if verbose: log.write("  -Warning: no data was retrieved. Skipping ...")
+            log.warning("No data was retrieved. Skipping ...", verbose=verbose)
             ref_genotype=dict()
             ref_genotype["variants/POS"]=np.array([],dtype="int64")
             return None
         
-        if verbose: log.write("  -Retrieving index...")
-        if verbose: log.write("  -Ref variants in the region: {}".format(len(ref_genotype["variants/POS"])))
+        log.write("  -Retrieving index...", verbose=verbose)
+        log.write("  -Ref variants in the region: {}".format(len(ref_genotype["variants/POS"])), verbose=verbose)
         #  match sumstats pos and ref pos: 
         # get ref index for its first appearance of sumstats pos
         #######################################################################################
@@ -170,7 +170,7 @@ def _get_rsq(    row,
             else:
                 # no position match
                 return None
-        if verbose: log.write("  -Matching variants using POS, NEA, EA ...")
+        log.write("  -Matching variants using POS, NEA, EA ...", verbose=verbose)
 
         sumstats["REFINDEX"] = sumstats.loc[:,["POS","NEA","EA"]].apply(lambda x: match_varaint(x), axis=1)
         log.write("  -Matched variants in sumstats and vcf:{} ".format(sum(~sumstats["REFINDEX"].isna())))
@@ -190,7 +190,7 @@ def _get_rsq(    row,
             lead_snp_genotype = GenotypeArray([ref_genotype["calldata/GT"][lead_snp_ref_index]]).to_n_alt()
             other_snp_genotype = GenotypeArray(ref_genotype["calldata/GT"][other_snps_ref_index]).to_n_alt()
             
-            if verbose: log.write("  -Calculating Rsq...")
+            log.write("  -Calculating Rsq...", verbose=verbose)
             
             if len(other_snp_genotype)>1:
                 valid_r2= np.power(rogers_huff_r_between(lead_snp_genotype,other_snp_genotype)[0],2)
@@ -198,7 +198,7 @@ def _get_rsq(    row,
                 valid_r2= np.power(rogers_huff_r_between(lead_snp_genotype,other_snp_genotype),2)
             sumstats.loc[~sumstats["REFINDEX"].isna(),"RSQ"] = valid_r2
         else:
-            if verbose: log.write("  -Lead SNP not found in reference...")
+            log.write("  -Lead SNP not found in reference...", verbose=verbose)
             sumstats["RSQ"]=None
         
         sumstats["RSQ"] = sumstats["RSQ"].astype("float")

gwaslab/util_ex_ldsc.py

@@ -0,0 +1,189 @@
+from gwaslab.ldsc_sumstats import estimate_h2
+from gwaslab.ldsc_sumstats import estimate_rg
+from gwaslab.g_Log import Log
+from gwaslab.qc_fix_sumstats import start_to
+from gwaslab.qc_fix_sumstats import finished
+from gwaslab.qc_fix_sumstats import skipped
+from gwaslab.io_read_ldsc import parse_ldsc_summary
+
+class ARGS():
+    def __init__(self, **args):
+        
+        self.out = "ldsc"
+        self.bfile = None
+        self.l2 = False
+        self.extract = None
+        self.keep = None
+        self.ld_wind_snps = None
+        self.ld_wind_kb = None
+        self.ld_wind_cm = None
+        self.print_snps = None
+        self.annot =None
+        self.thin_annot = False
+        self.cts_bin = None
+        self.cts_breaks = None
+        self.cts_names = None
+        self.per_allele = False
+        self.pq_exp =None
+        self.no_print_annot = False
+
+        if "h2" in args.keys():
+            self.h2 = args["h2"]
+        else:
+            self.h2 = None
+
+        self.h2_cts = None
+
+        if "rg" in args.keys():
+            self.rg = args["rg"]
+        else:
+            self.rg = None
+
+        if "ref_ld" in args.keys():
+            self.ref_ld = args["ref_ld"]
+        else:
+            self.ref_ld = None
+
+        if "ref_ld_chr" in args.keys():
+            self.ref_ld_chr = args["ref_ld_chr"]
+        else:
+            self.ref_ld_chr = None
+
+        if "w_ld" in args.keys():
+            self.w_ld = args["w_ld"]
+        else:
+            self.w_ld = None
+        
+        if "w_ld_chr" in args.keys():
+            self.w_ld_chr = args["w_ld_chr"]
+        else:
+            self.w_ld_chr = None     
+
+        self.overlap_annot = False
+        self.print_coefficients = "ldsc"
+        self.frqfile = None
+        self.frqfile_chr = None
+        self.no_intercept = None
+        self.intercept_h2 = None
+        self.intercept_gencov = None
+        self.M = None
+        self.two_step = None
+        self.chisq_max = None
+        self.ref_ld_chr_cts = None
+        self.print_cov = None
+        self.print_delete_vals = False
+        self.chunk_size = 50
+        self.pickle = False
+        self.yes_really = False
+        self.invert_anyway = False
+        self.n_blocks = 200
+        self.not_M_5_50 = False
+        self.no_check_alleles = False
+        self.return_silly_things = False
+
+        if "samp_prev" in args.keys():
+            self.samp_prev = args["samp_prev"]
+        else:
+            self.samp_prev = None 
+        
+        if "pop_prev" in args.keys():
+            self.pop_prev = args["pop_prev"]
+        else:
+            self.pop_prev = None 
+
+def _estimate_h2_by_ldsc(insumstats, log, verbose=True, **args):
+    sumstats = insumstats.copy()
+    ##start function with col checking##########################################################
+    _start_line = "run LD score regression"
+    _end_line = "running LD score regression"
+    _start_cols =[]
+    _start_function = ".estimate_h2_by_ldsc()"
+    _must_args ={}
+
+    is_enough_info = start_to(sumstats=sumstats,
+                            log=log,
+                            verbose=verbose,
+                            start_line=_start_line,
+                            end_line=_end_line,
+                            start_cols=_start_cols,
+                            start_function=_start_function,
+                            **_must_args)
+    if is_enough_info == False: return None
+    ############################################################################################
+    log.write(" -Run single variate LD score regression:", verbose=verbose)
+    log.write("  -Adopted from LDSC source code: https://github.com/bulik/ldsc", verbose=verbose)
+    log.write("  -Please cite LDSC: Bulik-Sullivan, et al. LD Score Regression Distinguishes Confounding from Polygenicity in Genome-Wide Association Studies. Nature Genetics, 2015.", verbose=verbose)
+    log.write(" -Arguments:", verbose=verbose)
+    
+    for key, value in args.items():
+        log.write("  -{}:{}".format(key, value), verbose=verbose)
+    
+    default_args = ARGS(**args)
+
+    if "Z" not in sumstats.columns:
+        sumstats["Z"] = sumstats["BETA"]/sumstats["SE"]
+
+    sumstats = sumstats.rename(columns={"EA":"A1","NEA":"A2","rsID":"SNP"})
+    
+    log.write(" -LDSC log:", verbose=verbose)
+    summary = estimate_h2(sumstats, default_args, log)
+    
+    log.write(" -Results have been stored in .ldsc_h2", verbose=verbose)
+    finished(log=log,verbose=verbose,end_line=_end_line)
+    return parse_ldsc_summary(summary)
+
+def _estimate_rg_by_ldsc(insumstats, other_traits ,log, verbose=True, **args):
+    sumstats = insumstats.copy()
+    ##start function with col checking##########################################################
+    _start_line = "run LD score regression for genetic correlation"
+    _end_line = "running LD score regression for genetic correlation"
+    _start_cols =[]
+    _start_function = ".estimate_rg_by_ldsc()"
+    _must_args ={}
+
+    is_enough_info = start_to(sumstats=sumstats,
+                            log=log,
+                            verbose=verbose,
+                            start_line=_start_line,
+                            end_line=_end_line,
+                            start_cols=_start_cols,
+                            start_function=_start_function,
+                            **_must_args)
+    if is_enough_info == False: return None
+    ############################################################################################
+    log.write(" -Run cross-trait LD score regression:", verbose=verbose)
+    log.write("  -Adopted from LDSC source code: https://github.com/bulik/ldsc", verbose=verbose)
+    log.write("  -Please cite LDSC: Bulik-Sullivan, B., et al. An Atlas of Genetic Correlations across Human Diseases and Traits. Nature Genetics, 2015.", verbose=verbose)
+    log.write(" -Arguments:", verbose=verbose)
+    
+    for key, value in args.items():
+        log.write("  -{}:{}".format(key, value), verbose=verbose)
+    
+    default_args = ARGS(**args)
+
+    if "Z" not in sumstats.columns:
+        sumstats["Z"] = sumstats["BETA"]/sumstats["SE"]
+
+    sumstats = sumstats.rename(columns={"EA":"A1","NEA":"A2","rsID":"SNP"})
+    
+    other_traits_to_use = []
+    alias = default_args.rg.split(",")[1:]
+
+    for index, each_other_sumstats in enumerate(other_traits):
+        log.write(" -Processing sumstats with alias {} ({})".format(alias[index], each_other_sumstats.meta["gwaslab"]["study_name"]))
+        if "rsID" not in each_other_sumstats.data.columns:
+            to_append = each_other_sumstats.filter_hapmap3(verbose=False).data.rename(columns={"EA":"A1","NEA":"A2","rsID":"SNP"})
+        else:        
+            to_append = each_other_sumstats.data.rename(columns={"EA":"A1","NEA":"A2","rsID":"SNP"})
+        
+        if "Z" not in to_append.columns:
+            to_append["Z"] = to_append["BETA"]/to_append["SE"]
+
+        other_traits_to_use.append(to_append[["SNP","A1","A2","Z","N"]])    
+
+    log.write(" -LDSC log:", verbose=verbose)
+    summary = estimate_rg(sumstats[["SNP","A1","A2","Z","N"]], other_traits_to_use, default_args, log)[1]
+    
+    log.write(" -Results have been stored in .ldsc_rg", verbose=verbose)
+    finished(log=log,verbose=verbose,end_line=_end_line)
+    return summary

gwaslab/util_in_calculate_gc.py

@@ -12,34 +12,34 @@ def lambdaGC(insumstats,include_chrXYMT=True, x=23 ,y=24, mt=25, mode="P",level=
     sumstats=insumstats.loc[:,["CHR",mode]]
     
     if include_chrXYMT is False:
-        if verbose: log.write(" -Excluding chrX, chrY, chrMT from lambda GC calculation.")
+        log.write(" -Excluding chrX, chrY, chrMT from lambda GC calculation.", verbose=verbose)
         xymt= [x,y,mt,"chrx","chry","chrmt","chrX","chrY","chrMT","chrM","M","x","y","mt","X","Y","MT"]
         sumstats = sumstats.loc[~sumstats["CHR"].isin(xymt),:]
 
     indata = sumstats[mode].values
     if len(indata) == 0: 
-        if verbose: log.write("  -No available variants to use for calculation.")
+        log.write("  -No available variants to use for calculation.", verbose=verbose)
         return np.nan   
     if mode=="p" or mode=="P":
         observedMedianChi2 = sp.stats.chi2.isf(np.nanmedian(indata),1)
         expectedMedianChi2 = sp.stats.chi2.ppf(level,1)
         lambdagc=observedMedianChi2/expectedMedianChi2
-        if verbose: log.write(" -Lambda GC (P mode) at "+ str(1 - level)+ " is"," ","{:.5f}".format(lambdagc))
+        log.write(" -Lambda GC (P mode) at "+ str(1 - level)+ " is"," ","{:.5f}".format(lambdagc), verbose=verbose)
     elif mode=="mlog10p" or mode=="MLOG10P":
         observedMedianChi2 = sp.stats.chi2.isf( np.nanmedian(np.power(10,-indata)) ,1)
         expectedMedianChi2 = sp.stats.chi2.ppf(level,1)
         lambdagc=observedMedianChi2/expectedMedianChi2
-        if verbose: log.write(" -Lambda GC (MLOG10P mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc))
+        log.write(" -Lambda GC (MLOG10P mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc), verbose=verbose)
     elif mode=="z" or mode=="Z":
         observedMedianChi2 = np.median((indata)**2)
         expectedMedianChi2 = sp.stats.chi2.ppf(level,1)
         lambdagc=observedMedianChi2/expectedMedianChi2
-        if verbose:log.write(" -Lambda GC (Z mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc))
+        if verbose:log.write(" -Lambda GC (Z mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc), verbose=verbose)
     elif mode=="chi2" or mode=="CHISQ":
         observedMedianChi2 = np.median(indata)
         expectedMedianChi2 = sp.stats.chi2.ppf(level,1)
         lambdagc=observedMedianChi2/expectedMedianChi2
-        if verbose:log.write(" -Lambda GC (CHISQ mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc))
+        log.write(" -Lambda GC (CHISQ mode) at "+ str(1- level)+ " is"," ","{:.5f}".format(lambdagc), verbose=verbose)
     else:
         return np.nan
     return lambdagc

gwaslab/util_in_calculate_power.py

@@ -21,24 +21,23 @@ def get_power(
               log=Log(),
               verbose=True
              ):
-    if verbose: log.write(" Start to calculate statistical power...")
+    log.write(" Start to calculate statistical power...", verbose=verbose)
     if mode=="b":
-        if verbose: 
-            log.write(" -Input settings (b mode):")
-            log.write("  -Number of cases:{}".format(ncase))
-            log.write("  -Number of controls:{}".format(ncontrol))
-            if genotype_rr is not None:
-                log.write("  -Risk allele RR:{:.3f}".format(genotype_rr))
-            elif genotype_or is not None:
-                log.write("  -Risk allele OR:{:.3f}".format(genotype_or))
-            elif beta is not None:
-                log.write("  -Risk allele beta:{:.3f}".format(beta))
-            else:
-                genotype_rr = 0.1
-                log.write("  -Risk allele RR:{:.3f}".format(genotype_rr))
-            log.write("  -Disease prevalence:{:.3f}".format(prevalence))
-            log.write("  -Risk allele frequency: {:.3f}".format(daf))
-            log.write("  -Significance level: {:.3e}".format(sig_level))
+        log.write(" -Input settings (b mode):", verbose=verbose)
+        log.write("  -Number of cases:{}".format(ncase), verbose=verbose)
+        log.write("  -Number of controls:{}".format(ncontrol), verbose=verbose)
+        if genotype_rr is not None:
+            log.write("  -Risk allele RR:{:.3f}".format(genotype_rr), verbose=verbose)
+        elif genotype_or is not None:
+            log.write("  -Risk allele OR:{:.3f}".format(genotype_or), verbose=verbose)
+        elif beta is not None:
+            log.write("  -Risk allele beta:{:.3f}".format(beta), verbose=verbose)
+        else:
+            genotype_rr = 0.1
+            log.write("  -Risk allele RR:{:.3f}".format(genotype_rr), verbose=verbose)
+        log.write("  -Disease prevalence:{:.3f}".format(prevalence), verbose=verbose)
+        log.write("  -Risk allele frequency: {:.3f}".format(daf), verbose=verbose)
+        log.write("  -Significance level: {:.3e}".format(sig_level), verbose=verbose)
         # Skol, A. D., Scott, L. J., Abecasis, G. R., & Boehnke, M. (2006). Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies. Nature genetics, 38(2), 209-213.
         aaf = daf**2
         abf = 2 * (daf) * (1 - daf)
@@ -56,11 +55,11 @@ def get_power(
                 # https://jamanetwork.com/journals/jama/fullarticle/188182
         
             if or_to_rr ==False:
-                if verbose: log.write(" -Alogorithm: Skol, Andrew D., et al. Nature genetics 38.2 (2006): 209-213....")
-                if verbose: log.write(" -GRR is approximated using OR. For prevalence < 10%, GRR is very similar to OR....")
+                log.write(" -Alogorithm: Skol, Andrew D., et al. Nature genetics 38.2 (2006): 209-213....", verbose=verbose)
+                log.write(" -GRR is approximated using OR. For prevalence < 10%, GRR is very similar to OR....", verbose=verbose)
             else:
-                if verbose: log.write(" -OR is converted to GRR using base prevalence: {}".format(prevalence))
-                if verbose: log.write(" -Alogorithm: Zhang, J., & Kai, F. Y. (1998). What's the relative risk?: A method of correcting the odds ratio in cohort studies of common outcomes. Jama, 280(19), 1690-1691.....")
+                log.write(" -OR is converted to GRR using base prevalence: {}".format(prevalence), verbose=verbose)
+                log.write(" -Alogorithm: Zhang, J., & Kai, F. Y. (1998). What's the relative risk?: A method of correcting the odds ratio in cohort studies of common outcomes. Jama, 280(19), 1690-1691.....", verbose=verbose)
             
         # additive
         x = [ 2*genotype_rr-1, genotype_rr, 1 ] 
@@ -68,19 +67,19 @@ def get_power(
         aap= x[0] * prevalence / (x[0]*aaf + x[1]*abf + x[2]*bbf)
         abp= x[1] * prevalence / (x[0]*aaf + x[1]*abf + x[2]*bbf)
         bbp= x[2] * prevalence / (x[0]*aaf + x[1]*abf + x[2]*bbf)
-        if verbose: log.write("Probability of disease :")
-        if verbose: log.write(" - Individuals with AA genotype: {:.3f}".format(aap))
-        if verbose: log.write(" - Individuals with AB genotype: {:.3f}".format(abp))
-        if verbose: log.write(" - Individuals with BB genotype: {:.3f}".format(bbp))
+        log.write("Probability of disease :", verbose=verbose)
+        log.write(" - Individuals with AA genotype: {:.3f}".format(aap), verbose=verbose)
+        log.write(" - Individuals with AB genotype: {:.3f}".format(abp), verbose=verbose)
+        log.write(" - Individuals with BB genotype: {:.3f}".format(bbp), verbose=verbose)
         
         pcase= (aap * aaf + abp * abf*0.5) / prevalence
         pcontrol=((1-aap )* aaf + (1-abp )* abf*0.5) / (1 - prevalence)
 
         vcase = pcase *(1-pcase)
         vcontrol =pcontrol *(1-pcontrol)
-        if verbose: log.write("Expected risk allele frequency:")
-        if verbose: log.write(" - In cases: {:.3f}".format(pcase))
-        if verbose: log.write(" - In controls: {:.3f}".format(pcontrol))
+        log.write("Expected risk allele frequency:", verbose=verbose)
+        log.write(" - In cases: {:.3f}".format(pcase), verbose=verbose)
+        log.write(" - In controls: {:.3f}".format(pcontrol), verbose=verbose)
 
         num= (pcase - pcontrol)
         den= np.sqrt( (vcase/ncase +  vcontrol/ncontrol)*0.5 )
@@ -88,22 +87,22 @@ def get_power(
 
         c = ss.norm.isf(sig_level/2)
         power = 1 - ss.norm.cdf(c-u) + ss.norm.cdf(-c-u)
-        if verbose: log.write("Expected power: {:.3f}".format(power))
+        log.write("Expected power: {:.3f}".format(power), verbose=verbose)
 
     elif mode=="q":
         if beta is None:
             beta = 0.1
-        if verbose:
-            log.write(" -Input settings (q mode):")
-            log.write("  -Significance level: {}".format(sig_level))
-            log.write("  -EAF: {}".format(eaf))
-            log.write("  -BETA: {}".format(beta))
-            log.write("  -N: {}".format(n))
-            log.write("  -SNPR2: {}".format(2*eaf*(1-eaf)*(beta**2)))
+
+        log.write(" -Input settings (q mode):", verbose=verbose)
+        log.write("  -Significance level: {}".format(sig_level), verbose=verbose)
+        log.write("  -EAF: {}".format(eaf), verbose=verbose)
+        log.write("  -BETA: {}".format(beta), verbose=verbose)
+        log.write("  -N: {}".format(n), verbose=verbose)
+        log.write("  -SNPR2: {}".format(2*eaf*(1-eaf)*(beta**2)), verbose=verbose)
         c = ss.chi2.isf(sig_level,df=1)
         NCP = n * 2*eaf*(1-eaf)*(beta**2)/vary
         power = 1 - ss.ncx2.cdf(c, df=1, nc=NCP)
-    if verbose: log.write("Finished calculating statistical power.")
+    log.write("Finished calculating statistical power.", verbose=verbose)
     return power
 
 def get_beta(
@@ -137,11 +136,11 @@ def get_beta(
             eafs = np.linspace(eaf_range[1],eaf_range[0],n_matrix)
             betas =  np.linspace(beta_range[0],beta_range[1],n_matrix)
             
-            if verbose: log.write(" -Updating eaf-beta matrix...")
+            log.write(" -Updating eaf-beta matrix...", verbose=verbose)
             for i in range(n_matrix):
                     eaf_beta_matrix[i,] = calculate_power_single(beta=betas,eaf=eafs[i],n=n,sig_level=sig_level,vary=vary)
             
-            if verbose: log.write(" -Extracting eaf-beta combinations with power = {}...".format(t))
+            log.write(" -Extracting eaf-beta combinations with power = {}...".format(t), verbose=verbose)
             i,j=1,1
             eaf_beta = []
             while i<n_matrix-1 and j<n_matrix-1:
@@ -207,11 +206,11 @@ def get_beta_binary(
         eafs = np.linspace(eaf_range[1],eaf_range[0],n_matrix)
         betas =  np.linspace(beta_range[0],beta_range[1],n_matrix)
         
-        if verbose: log.write(" -Updating eaf-beta matrix...")
+        log.write(" -Updating eaf-beta matrix...", verbose=verbose)
         if or_to_rr ==False:
-            if verbose: log.write(" -GRR is approximated using OR. For prevalence < 10%, GRR is very similar to OR....")
+            log.write(" -GRR is approximated using OR. For prevalence < 10%, GRR is very similar to OR....", verbose=verbose)
         else:
-            if verbose: log.write(" -OR is converted to GRR using base prevalence: {}".format(prevalence))
+            log.write(" -OR is converted to GRR using base prevalence: {}".format(prevalence), verbose=verbose)
         
         for i in range(n_matrix):
                 eaf_beta_matrix[i,] = calculate_power_single(beta=betas,
@@ -222,7 +221,7 @@ def get_beta_binary(
                                                                 sig_level=sig_level,
                                                                 or_to_rr=or_to_rr)
         
-        if verbose: log.write(" -Extracting eaf-beta combinations with power = {}...".format(t))
+        log.write(" -Extracting eaf-beta combinations with power = {}...".format(t), verbose=verbose)
         i,j=1,1
         eaf_beta = []
         while i<n_matrix-1 and j<n_matrix-1:

gwaslab/util_in_convert_h2.py

@@ -65,7 +65,7 @@ def _get_per_snp_r2(sumstats,
            adjuested=False,
            verbose=True):
     # Pierce, B. L., Ahsan, H., & VanderWeele, T. J. (2011). Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. International journal of epidemiology, 40(3), 740-752.
-    if verbose: log.write("Start to calculate per-SNP heritibility...")
+    log.write("Start to calculate per-SNP heritibility...", verbose=verbose)
     if type(k) is int or type(k) is float:
        pass 
     elif k =="all":
@@ -81,18 +81,18 @@ def _get_per_snp_r2(sumstats,
             # Var(e) = betase**2 * 2 * N * MAF * (1-MAF)
             # r2 = Var(beta * X) / Var(y)
 
-            if verbose: log.write(" -Calculating per-SNP rsq by 2 * (BETA**2) * AF * (1-AF) / Var(y)...")
+            log.write(" -Calculating per-SNP rsq by 2 * (BETA**2) * AF * (1-AF) / Var(y)...", verbose=verbose)
             sumstats["_VAR(BETAX)"] = 2*(sumstats[beta]**2)*sumstats[af]*(1-sumstats[af])
             
             if type(vary) is int or type(vary) is float:
-                if verbose: log.write(" -Var(y) is provided: {}...".format(vary))
+                log.write(" -Var(y) is provided: {}...".format(vary), verbose=verbose)
                 sumstats["SNPR2"] = sumstats["_VAR(BETAX)"] / vary
             elif vary=="se":
-                if verbose: log.write(" -Var(y) is estimated from VAR(BETA * X), N, MAF, SE: {}...".format(vary))
+                log.write(" -Var(y) is estimated from VAR(BETA * X), N, MAF, SE: {}...".format(vary), verbose=verbose)
                 sumstats["_SIGMA2"] = sumstats[se]**2 * 2*(sumstats[n])*sumstats[af]*(1-sumstats[af])
                 sumstats["SNPR2"] = sumstats["_VAR(BETAX)"] / (sumstats["_SIGMA2"] + sumstats["_VAR(BETAX)"]) 
         else:
-            if verbose: log.write(" -Warning: Not enough informationfor calculation.")
+            log.warning("Not enough information for calculation.")
     
     if mode=="b":
         if ncase not in sumstats.columns:
@@ -117,11 +117,11 @@ def _get_per_snp_r2(sumstats,
     else:
         snpr2 = "SNPR2"
     if n in sumstats.columns:
-        if verbose: log.write(" -Calculating F-statistic: F = [(N-k-1)/k] * (r2/1-r2)... where k = {}".format(k))
-        if verbose: log.write(" -For r2, {} is used.".format(snpr2))
+        log.write(" -Calculating F-statistic: F = [(N-k-1)/k] * (r2/1-r2)... where k = {}".format(k), verbose=verbose)
+        log.write(" -For r2, {} is used.".format(snpr2), verbose=verbose)
         sumstats["F"] = sumstats[snpr2]*(sumstats[n]-1 -k)/((1-sumstats[snpr2]) * k)
         
-    if verbose: log.write("Finished calculating per-SNP heritability!")
+    log.write("Finished calculating per-SNP heritability!", verbose=verbose)
     return sumstats
 #
 def get_population_allele_frequency(af, prop, odds_ratio, prevalence,eps=1e-15):