gwaslab 3.6.2__py3-none-any.whl → 3.6.4__py3-none-any.whl

gwaslab/bd_get_hapmap3.py

@@ -42,14 +42,39 @@ def gethapmap3(sumstats,rsid="rsID",chrom="CHR", pos="POS", ea="EA", nea="NEA",b
         additional_cols= ["A1","A2"]
     else:
         additional_cols=[]
+    
     hapmap3_ref = pd.read_csv(data_path,sep="\s+",usecols=["#CHROM","POS","rsid"]+additional_cols, dtype={"#CHROM":"string","POS":"string"})
+    
     #rsid    A1      A2      #CHROM  POS
     #rs3094315       G       A       1       752566
     
     if rsid in sumstats.columns:
         log.write(" -rsID will be used for matching...", verbose=verbose)
-        output = sumstats.loc[sumstats[rsid].isin(hapmap3_ref["rsid"].values),:].copy()
-        log.write(" -Raw input contains "+str(len(output))+" Hapmap3 variants based on rsID...", verbose=verbose)
+        hapmap3_ref = hapmap3_ref.rename(columns={"rsid":rsid})
+        
+        output = sumstats.loc[sumstats[rsid].isin(hapmap3_ref[rsid].values),:].copy()
+        
+        output = pd.merge(output, hapmap3_ref, on = rsid, how=how, suffixes=('', '_hapmap3'))
+
+        raw_rsid_count= len(output)
+        log.write(f" -Raw input contains {raw_rsid_count} Hapmap3 variants based on rsID...", verbose=verbose)
+
+        if match_allele:
+            log.write(" -Checking if alleles are same...")
+            is_matched = ((output[ea].astype("string") == output["A1"]) & (output[nea].astype("string") == output["A2"])) \
+                            | ((output[ea].astype("string") == output["A2"]) & (output[nea].astype("string") == output["A1"]))
+            if how=="right":
+                is_matched = ((output[ea].astype("string") == output["A1"]) & (output[nea].astype("string") == output["A2"])) \
+                            | ((output[ea].astype("string") == output["A2"]) & (output[nea].astype("string") == output["A1"])) | output[ea].isna()
+            output = output.loc[is_matched,:]
+            output = output.drop(columns=["#CHROM","A1","A2"] )
+            log.write(f" -Filtered {raw_rsid_count - len(output)} Hapmap3 variants due to unmatech alleles...", verbose=verbose)
+        
+        for i in ["#CHROM","A1","A2","POS_hapmap3"]:
+            todrop=[]
+            if i in output.columns:
+                todrop.append(i)
+        output = output.drop(columns=todrop)
         return output
     
     elif chrom in sumstats.columns and pos in sumstats.columns:

gwaslab/g_Sumstats.py

@@ -102,7 +102,7 @@ from gwaslab.util_ex_run_scdrs import _run_scdrs
 #20220309
 class Sumstats():
     def __init__(self,
-             sumstats,
+             sumstats=None,
              fmt=None,
              tab_fmt="tsv",
              snpid=None,
@@ -144,10 +144,11 @@ class Sumstats():
              dof=None,
              snpr2=None,
              status=None,
-             other=[],
+             other=None,
+             exclude=None,
+             include=None,
              chrom_pat=None,
              snpid_pat=None,
-             usekeys=None,
              direction=None,
              verbose=True,
              study="Study_1",
@@ -238,7 +239,8 @@ class Sumstats():
           trait=trait,
           status=status,
           other=other,
-          usekeys=usekeys,
+          exclude=exclude,
+          include=include,
           chrom_pat=chrom_pat,
           snpid_pat=snpid_pat,
           verbose=verbose,
@@ -912,10 +914,13 @@ class Sumstats():
                                                                   verbose=verbose, 
                                                                   **kwargs)
     
-    def estimate_rg_by_ldsc(self, build=None, verbose=True, match_allele=True, how="right",**kwargs):
+    def estimate_rg_by_ldsc(self, build=None, verbose=True, match_allele=True, how="right", get_hm3=True,**kwargs):
         if build is None:
             build = self.meta["gwaslab"]["genome_build"]
-        insumstats = gethapmap3(self.data.copy(), build=build, verbose=verbose , match_allele=match_allele, how=how )
+        if get_hm3==True:
+            insumstats = gethapmap3(self.data.copy(), build=build, verbose=verbose , match_allele=match_allele, how=how )
+        else:
+            insumstats = self.data
         ldsc_rg = _estimate_rg_by_ldsc(insumstats=insumstats,
                                              meta=self.meta,
                                              log=self.log, 

gwaslab/g_SumstatsMulti.py

@@ -53,6 +53,7 @@ class SumstatsMulti( ):
                  build="99",
                  engine="pandas",
                  merge_mode="outer",
+                 merge_by_id=False,
                  verbose=True ):
         
         for i,sumstatsObject in enumerate(sumstatsObjects):
@@ -152,7 +153,7 @@ class SumstatsMulti( ):
             for i, sumstatsObject in enumerate(sumstatsObjects):
                 if i >0:
                     self.log.write("Merging Sumstats #{} to main DataFrame...".format(i+1))
-                    self.data = self._merge_two_sumstats(sumstatsObject.data,i=i,merge_mode=merge_mode,engine=engine)
+                    self.data = self._merge_two_sumstats(sumstatsObject.data,i=i,merge_mode=merge_mode,engine=engine,merge_by_id=merge_by_id)
                     self.log.write("Finished merging Sumstats #{} to main DataFrame.".format(i+1))
         else:
             for i, sumstatsObject in enumerate(sumstatsObjects):
@@ -167,6 +168,7 @@ class SumstatsMulti( ):
                             verbose=True,
                             merge_mode="outer",
                             engine="pandas",
+                            merge_by_id=False,
                             i=0):
 
         # _1 _2 
@@ -184,6 +186,7 @@ class SumstatsMulti( ):
                                                         log=self.log,
                                                         verbose=verbose,
                                                         merge_mode=merge_mode,
+                                                        merge_by_id=merge_by_id,
                                                         stats_cols1 = self.other_cols[0],
                                                         stats_cols2 = self.other_cols[i],
                                                         suffixes=("_1",""),

gwaslab/g_version.py

@@ -2,11 +2,13 @@ from gwaslab.g_Log import Log
 import subprocess
 import os
 import numpy as np 
+import sys
 
 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) 2022-2025, Yunye He, Kamatani Lab, GPL-3.0 license, gwaslab@gmail.com",verbose=verbose)
+    log.write(f"Python version: {sys.version}",verbose=verbose)
 
 def _get_version():
     # return short version string like v3.4.33
@@ -15,8 +17,8 @@ def _get_version():
 def gwaslab_info():
     # version meta information
     dic={
-   "version":"3.6.2",
-   "release_date":"20250509"
+   "version":"3.6.4",
+   "release_date":"20250624"
     }
     return dic   
 

gwaslab/hm_casting_polars.py

@@ -18,30 +18,95 @@ def _merge_mold_with_sumstats_by_chrposp(mold, sumstats, ref_path=None,add_raw_i
                                         log=Log(),
                                         suffixes=("_MOLD",""),
                                         merge_mode="full",
+                                        merge_by_id=False,
                                         verbose=True,
                                         return_not_matched_mold =False):
     
     log.write("Start to merge sumstats...", verbose=verbose)
     if merge_mode=="full":
+        
         sumstats = sumstats.rename({
                                             "SNPID":"_SNPID_RIGHT",
                                             "rsID":"_rsID_RIGHT"
                                             }, strict=False) #,
+
     
-    # drop old ids
-    cols_to_drop = []
-    for i in sumstats.columns:
-        if i in ["SNPID","rsID"]:
-            cols_to_drop.append(i)    
-    if len(cols_to_drop)>0:
-        log.write(" -Dropping old IDs:{}".format(cols_to_drop), verbose=verbose)
-        sumstats = sumstats.drop(columns=cols_to_drop)
+    if merge_by_id==False:    
+        # drop old ids
+        cols_to_drop = []
+        for i in sumstats.columns:
+            if i in ["SNPID","rsID"]:
+                cols_to_drop.append(i)    
+        if len(cols_to_drop)>0:
+            log.write(" -Dropping old IDs:{}".format(cols_to_drop), verbose=verbose)
+            sumstats = sumstats.drop(columns=cols_to_drop)
+
+        ##################################################################################################################
+
+        # mold sumffix + mold 
+            # add ASET
+        mold = mold.with_columns(
+                pl.when( pl.col("EA_1") > pl.col("NEA_1") )  
+                .then(   pl.col("EA_1") + ":" + pl.col("NEA_1")  )  
+                .otherwise( pl.col("NEA_1") + ":" + pl.col("EA_1") )
+                .alias("ASET")
+            )
+        
+        sumstats = sumstats.with_columns(
+                pl.when( pl.col("EA") > pl.col("NEA") )  
+                .then(   pl.col("EA") + ":" + pl.col("NEA")  )  
+                .otherwise( pl.col("NEA") + ":" + pl.col("EA") )
+                .alias("ASET"))
+        
+        sumstats_len = len(sumstats)
+        mold_len = len(mold)
+        sumstats = sumstats.unique(subset=["CHR","POS","ASET"]) 
+        mold = mold.unique(subset=["CHR","POS","ASET"]) 
+
+        log.write(f' -Left:  dropping duplicated variants based on CHR,POS,ASET: {sumstats_len - len(sumstats)}')
+        log.write(f' -Right: dropping duplicated variants based on CHR,POS,ASET: {mold_len - len(mold)}')
 
-    ##################################################################################################################
+        mold = mold.with_columns(
+            pl.when( pl.col("NEA_1").str.len_chars() != pl.col("EA_1").str.len_chars() )  
+            .then(   
+                pl.when( pl.col("EAF_1")<0.5 ).then(
+                    pl.col("ASET") + ":" + pl.col("EA_1") 
+                ).otherwise( pl.col("ASET") + ":" + pl.col("NEA_1") )
+                .alias("ASET")
+                )  
+            .otherwise( pl.col("ASET") )
+            .alias("ASET")
+        )
+        
+        sumstats = sumstats.with_columns(
+            pl.when( pl.col("NEA").str.len_chars() != pl.col("EA").str.len_chars() )  
+            .then(   
+                pl.when( pl.col("EAF")<0.5 ).then(
+                    pl.col("ASET") + ":" + pl.col("EA") 
+                ).otherwise( pl.col("ASET") + ":" + pl.col("NEA") )
+                .alias("ASET")
+                )  
+            .otherwise( pl.col("ASET"))
+            .alias("ASET")
+            )
+
+        mold_sumstats = mold.join(sumstats, on=["CHR","POS","ASET"], how=merge_mode, suffix="_", coalesce=True)
+    
+    elif merge_by_id==True:
+
+        sumstats = sumstats.rename({
+                                            "_SNPID_RIGHT":"SNPID",
+                                            }, strict=False)
+        
 
-    # mold sumffix + mold 
+        sumstats_len = len(sumstats)
+        mold_len = len(mold)
+        sumstats = sumstats.unique(subset=["SNPID","CHR","POS"])
+        mold = mold.unique(subset=["SNPID","CHR","POS"]) 
+        log.write(f' -Left:  dropping duplicated variants based on CHR,POS,SNPID: {sumstats_len - len(sumstats)}')
+        log.write(f' -Right: dropping duplicated variants based on CHR,POS,SNPID: {mold_len - len(mold)}')
+        mold_sumstats = mold.join(sumstats, on=["SNPID","CHR","POS"], how=merge_mode, suffix="_", coalesce=True)
 
-    mold_sumstats = mold.join(sumstats, on=["CHR","POS"], how=merge_mode, suffix="_", coalesce=True)
 
     if merge_mode=="full":
         is_temp_na = mold_sumstats["EA_1"].is_null()
@@ -50,13 +115,17 @@ def _merge_mold_with_sumstats_by_chrposp(mold, sumstats, ref_path=None,add_raw_i
         for i in ["EA_1","NEA_1","EA","NEA"]:
             mold_sumstats = mold_sumstats.with_columns(pl.col(i).cast(pl.String).alias(i))
 
-        # for variants not in template, copy snp info
-        mold_sumstats = mold_sumstats.with_columns(
+        if merge_by_id==False:
+            mold_sumstats = mold_sumstats.with_columns(
             pl.when( is_temp_na )  
                 .then(   pl.col("_SNPID_RIGHT")  )  
                 .otherwise( pl.col("SNPID") )
                 .alias("SNPID")
-        ).with_columns(
+            )
+            mold_sumstats = mold_sumstats.drop(["_SNPID_RIGHT"])
+
+        # for variants not in template, copy snp info
+        mold_sumstats = mold_sumstats.with_columns(
             pl.when( is_temp_na )  
                 .then( pl.col("EA")  )  
                 .otherwise( pl.col("EA_1") )
@@ -66,12 +135,18 @@ def _merge_mold_with_sumstats_by_chrposp(mold, sumstats, ref_path=None,add_raw_i
                 .then( pl.col("NEA")  )  
                 .otherwise( pl.col("NEA_1") )
                 .alias("NEA_1")
+        ).with_columns(
+            pl.when( is_temp_na )  
+                .then( pl.col("EAF")  )  
+                .otherwise( pl.col("EAF_1"))
+                .alias("EAF_1")
         ).with_columns(
             pl.when( is_temp_na )  
                 .then( pl.col("STATUS")  )  
                 .otherwise( pl.col("STATUS_1") )
                 .alias("STATUS_1")
         )
+
         # 
         if "_rsID_RIGHT" in mold_sumstats.columns:
             mold_sumstats = mold_sumstats.with_columns(
@@ -87,7 +162,7 @@ def _merge_mold_with_sumstats_by_chrposp(mold, sumstats, ref_path=None,add_raw_i
         
         mold_sumstats = mold_sumstats.with_columns(
                 pl.when( is_temp_na_2 )  
-                .then(   pl.col("EA_1")  )  
+                .then(   pl.col("EA_1") )  
                 .otherwise( pl.col("EA") )
                 .alias("EA")
                 ).with_columns(
@@ -97,10 +172,14 @@ def _merge_mold_with_sumstats_by_chrposp(mold, sumstats, ref_path=None,add_raw_i
                 .alias("NEA")
                 )
         
-        mold_sumstats = mold_sumstats.drop(["_SNPID_RIGHT"])
+        
+    if merge_by_id==False:
+        mold_sumstats = mold_sumstats.unique(subset=["CHR","POS","ASET"])
+        log.write(" -After merging by CHR, POS and ASET:{}".format(len(mold_sumstats)), verbose=verbose)
+    else:
+        mold_sumstats = mold_sumstats.unique(subset=["SNPID","CHR","POS"])
+        log.write(" -After merging by SNPID, CHR and POS:{}".format(len(mold_sumstats)), verbose=verbose)
 
-    log.write(" -After merging by CHR and POS:{}".format(len(mold_sumstats)), verbose=verbose)
-    
     mold_sumstats = _keep_variants_with_same_allele_setp(mold_sumstats,suffixes=suffixes)
 
     log.write(" -Matched variants:{}".format(len(mold_sumstats)), verbose=verbose)
@@ -116,13 +195,9 @@ def _keep_variants_with_same_allele_setp(sumstats, log=Log(),verbose=True,suffix
     
     is_perfect_match = (sumstats[ea2] == sumstats[ea1]) & (sumstats[nea2] == sumstats[nea1])
     is_flipped_match = (sumstats[ea2] == sumstats[nea1]) & (sumstats[nea2] == sumstats[ea1])
-    is_allele_set_match = is_flipped_match | is_perfect_match
     
-    log.write(" -Matching alleles and keeping only variants with same allele set: ", verbose=verbose)
     log.write("  -Perfect match: {}".format(sum(is_perfect_match)), verbose=verbose)
     log.write("  -Flipped match: {}".format(sum(is_flipped_match)), verbose=verbose)
-    log.write("  -Unmatched : {}".format(sum(~is_allele_set_match)), verbose=verbose)
-    sumstats = sumstats.filter(is_allele_set_match)
     return sumstats
 
 def _align_with_moldp(sumstats, log=Log(),verbose=True, suffixes=("_MOLD","")):

gwaslab/io_preformat_input.py

@@ -58,8 +58,9 @@ def preformat(sumstats,
           study=None,
           trait=None,
           build=None,
-          other=[],
-          usekeys=None,
+          other=None,
+          exclude=None,
+          include=None,
           chrom_pat=None,
           snpid_pat=None,
           verbose=False,
@@ -68,15 +69,22 @@ def preformat(sumstats,
 
     #renaming dictionary
     rename_dictionary = {}
-    usecols = []
-    dtype_dictionary ={}    
+    usecols = list()
+    if other is None:
+        other = list()
+    if exclude is None:
+        exclude = list()
+    if include is None:
+        include = list()
+    
+    dtype_dictionary = {}    
     if readargs is None:
         readargs={}
  #######################################################################################################################################################
     # workflow: 
     # 1. formatbook
     # 2. user specified header
-    # 3. usekeys
+    # 3. include & exclude
     if tab_fmt=="parquet":
         if type(sumstats) is str:
             log.write("Start to load data from parquet file....",verbose=verbose)
@@ -316,18 +324,36 @@ def preformat(sumstats,
             study = raw_cols[9]
             usecols =  usecols + [study]
 
-    if usekeys is not None:
+    if len(include)>0:
     # extract only specified keys
         usecols_new =[]
-        for i in usekeys:
+        for i in include:
+            # rename_dictionary: sumstats to gwaslab
             for k, v in rename_dictionary.items():
                 if i == v:
+                    # get list of sumstats header
                     usecols_new.append(k)
         usecols_valid =[]
         for i in usecols_new:
             if i in usecols:
                 usecols_valid.append(i)
+        log.write(f' -Include columns :{",".join(usecols_valid)}' ,verbose=verbose)
         usecols = usecols_valid
+
+    if len(exclude)>0:
+    # exclude specified keys
+        exclude_cols =[]
+        for i in exclude:
+            # rename_dictionary: sumstats to gwaslab
+            for k, v in rename_dictionary.items():
+                if i == v:
+                    # get list of sumstats header
+                    exclude_cols.append(k)
+        log.write(f' -Exclude columns :{",".join(exclude_cols)}' ,verbose=verbose)
+        for i in exclude_cols:
+            if i in usecols:
+                usecols.remove(i)
+
  #loading data ##########################################################################################################
     
     try:

gwaslab/qc_fix_sumstats.py

@@ -107,6 +107,7 @@ def fixID(sumstats,
         except:
             log.write(" -Force converting rsID to pd.string data type...",verbose=verbose)
             sumstats[rsid] = sumstats[rsid].astype("string")
+    
     if snpid in sumstats.columns: 
         # convert to string datatype
         try:
@@ -147,6 +148,7 @@ def fixID(sumstats,
         log.write(" -A look at the unrecognized rsID :",set(sumstats.loc[(~is_rsid)&(~is_rs_chrpos),rsid].head()),"...", verbose=verbose) 
       
     ############################  fixing chr pos###################################################  
+    
     if fixchrpos == True:
     # from snpid or rsid, extract CHR:POS to fix CHR and POS    
         if snpid in sumstats.columns: 

gwaslab/util_ex_run_clumping.py

@@ -9,7 +9,7 @@ from gwaslab.util_ex_process_ref import _process_plink_input_files
 from gwaslab.g_version import _checking_plink_version
 
 def _clump(insumstats, vcf=None, scaled=False, out="clumping_plink2", 
-           p="P",mlog10p="MLOG10P", overwrite=False, study=None, bfile=None, 
+           p="P",mlog10p="MLOG10P", overwrite=False, study=None, bfile=None, pfile=None,
            n_cores=1, memory=None, chrom=None, clump_p1=5e-8, clump_p2=5e-8, clump_r2=0.01, clump_kb=250,
            log=Log(),verbose=True,plink="plink",plink2="plink2"):
     ##start function with col checking##########################################################
@@ -54,6 +54,7 @@ def _clump(insumstats, vcf=None, scaled=False, out="clumping_plink2",
     # process reference file
     bfile, plink_log, ref_bim,filetype = _process_plink_input_files(chrlist=sumstats["CHR"].unique(), 
                                                        bfile=bfile, 
+                                                       pfile=pfile,
                                                        vcf=vcf, 
                                                        n_cores=n_cores,
                                                        plink_log=plink_log, 
@@ -162,9 +163,12 @@ def _clump(insumstats, vcf=None, scaled=False, out="clumping_plink2",
             log.write(e.output)
         #os.system(script)
         
-        clumped = pd.read_csv("{}.clumps".format(out_single_chr),sep="\s+")
-        results = pd.concat([results,clumped],ignore_index=True)
-        
+        try:
+            clumped = pd.read_csv("{}.clumps".format(out_single_chr),sep="\s+")
+            results = pd.concat([results,clumped],ignore_index=True)
+            
+        except:
+            log.write(f"Clumping failed for chr{i}")
         # remove temp SNPIDP file 
         os.remove(clump)
     

gwaslab/util_ex_run_scdrs.py

@@ -84,7 +84,7 @@ def _run_scdrs( scdrs="scdrs",
         bash_script+=f'''
 {python} {scdrs} perform-downstream {analysis_string} \
     --h5ad-file {h5ad_file} \
-    --score-file {trait}.full_score.gz \
+    --score-file ./{trait}.full_score.gz \
     --out-folder {out_folder} \
     --min_genes 250 \
     --min_cells 50 \

gwaslab/viz_aux_quickfix.py

@@ -199,7 +199,6 @@ def _quick_assign_i_with_rank(sumstats, chrpad, use_rank=False, chrom="CHR",pos=
             posdiccul[i]= posdiccul[i-1] + posdiccul[i] + sumstats[pos].max()*chrpad
     else:
         posdiccul = _posdiccul
-
     # convert base pair postion to x axis position using the cumulative sum dictionary
     sumstats["_ADD"]=sumstats[chrom].apply(lambda x : posdiccul[int(x)-1])
     

gwaslab/viz_plot_compare_effect.py

@@ -985,6 +985,7 @@ def test_q(df,beta1,se1,beta2,se2,q_level=0.05,is_q_mc=False, log=Log(), verbose
         log.write(" -Bonferroni correction applied...", verbose=verbose)
         df[rawpq] = df[pq] 
         df[pq] = df[pq] * len(df[pq])
+        df.loc[df[pq]>1,pq] = 1
 
     df.loc[df[pq]<q_level,"Edge_color"]="black"
     df.drop(columns=["Weight_1","Weight_2","BETA_FE"],inplace=True)
@@ -1330,17 +1331,20 @@ def configure_legend(fig, ax, legend_mode, is_q, is_q_mc, legend_elements, legen
     #    handle.set_edgecolor("white")
 
     ## Move titles to the left 
-    for item, label in zip(L.legendHandles, L.texts):
-        if label._text  in legend_elements:
-            item.set_edgecolor("white")
-            #item._legmarker.set_markersize(scatterargs["s"]*1.5)
-            item._sizes = [scatterargs["s"]*2]
-        if legend_mode == "full":
-            if label._text  in [legend_title, legend_title2]:
-                width=item.get_window_extent(fig.canvas.get_renderer()).width
-                label.set_ha('left')
-                label.set_position((-8*width,0))
-
+    try:
+        for item, label in zip(L.legendHandles, L.texts):
+            if label._text  in legend_elements:
+                item.set_edgecolor("white")
+                #item._legmarker.set_markersize(scatterargs["s"]*1.5)
+                item._sizes = [scatterargs["s"]*2]
+            if legend_mode == "full":
+                if label._text  in [legend_title, legend_title2]:
+                    width=item.get_window_extent(fig.canvas.get_renderer()).width
+                    label.set_ha('left')
+                    label.set_position((-8*width,0))
+    except:
+        pass
+    
     ax.tick_params(axis='both', labelsize=font_kwargs["fontsize"])
     plt.setp(L.texts,**font_kwargs)
     plt.setp(L.get_title(),**font_kwargs)

gwaslab/viz_plot_miamiplot2.py

@@ -70,8 +70,9 @@ def plot_miami2(
           region_hspace = 0.1,
           dpi=100,
           fontsize = 10,
+          xtick_label_size = 10,
           font_family="Arial",
-          xlabel_coords=(-0.01, -0.027),
+          xlabel_coords=None,
           xtick_label_pad=None,
           verbose=True,
           xtickpad=None,
@@ -142,9 +143,12 @@ def plot_miami2(
         scatter_args={}
 
     if fig_args is None:
-        fig_args= {"figsize":(15,5),"dpi":100}
+        fig_args= {"figsize":(15,10),"dpi":100}
     if save_args is None:
         save_args={"dpi":100,"facecolor":"white"}
+    
+    if xlabel_coords is None:
+        xlabel_coords = (-0.01,- region_hspace/2 )
 
     # figure out mqq args
     mqq_args1,mqq_args2 = _sort_args_to_12(mqq_args)
@@ -157,19 +161,10 @@ def plot_miami2(
 
     if dpi!=100:
         fig_args["dpi"] = dpi
-    if xtickpad is None:
-        if "figsize" not in fig_args.keys():
-            fig_args["figsize"] = (15,5)
-        xtickpad =   fig_args["figsize"][1] * region_hspace *72 / 6
-    if xtick_label_pad is None:
-        if "figsize" not in fig_args.keys():
-            fig_args["figsize"] = (15,5)
-        xtick_label_pad =  72 * fig_args["figsize"][1] * region_hspace / 6
     
     if titles is None:
         titles=["",""]
     
-    
     titles_pad_adjusted=[1,0]
     if titles_pad is None:
         titles_pad=[0.2,0.2]
@@ -242,11 +237,12 @@ def plot_miami2(
     # P_1  scaled_P_1  P_2  scaled_P_2  TCHR+POS CHR POS    
     log.write(" -Columns in merged sumstats: {}".format(",".join(merged_sumstats.columns)), verbose=verbose)
 
-
     del(sumstats1)
     del(sumstats2)
     garbage_collect.collect()
     #####################################################################################################################
+
+
     ##plotting
     if figax is None:
         #fig_args["figsize"] = (15,10)
@@ -254,7 +250,30 @@ def plot_miami2(
         plt.subplots_adjust(hspace=region_hspace)    
     else:
         fig, ax1, ax5 = figax
+    ##########################################################################################################################
+    bbox1 = ax1.get_position() 
+    bbox5 = ax5.get_position()
+    
 
+
+    fig_height_inches = fig.get_figheight()
+    ax_height_inches = (bbox1.height + bbox5.height) * fig_height_inches /2
+    ax_height_points = ax_height_inches * 72
+    tick = ax1.xaxis.get_major_ticks()[0].tick1line
+    # Tick length is determined by its 'markersize' (in points)
+    tick_length = tick.get_markersize()
+    
+    tick_points_to_pixels = tick_length * fig.dpi / 72.0
+    ax_height_pixels = bbox1.height * fig.get_figheight() * fig.dpi
+    tick_axes_length = tick_points_to_pixels / ax_height_pixels
+
+    if xtick_label_pad is None:
+        if "figsize" not in fig_args.keys():
+            fig_args["figsize"] = (15,10)
+        # (total hsapce - tick label font size) / 2 
+        xtick_label_pad = 0
+        #xtick_label_pad =  ((ax_height_points * region_hspace) - 2*tick_length - xtick_label_size) / 2 
+    ########################################################################################################################
     #if same_ylim==True:
         #maxy = merged_sumstats[["scaled_P_1","scaled_P_2"]].max().max()
 
@@ -307,17 +326,23 @@ def plot_miami2(
     #ax5.set_xticks(chrom_df)
     ax5.set_xticklabels([])
     ax5.xaxis.set_ticks_position("top")
+    ax5.tick_params(axis='x', which='major', pad=0)
 
     # Ad#just the visibility for spines #######################################################
     ax1, ax5 = _set_spine_visibility(ax1, ax5)
     ######################################################################################################################
 #####################################################################################################################
     # set labels
-    ax1.set_xlabel("Chromosome",fontsize=fontsize,family=font_family)
+    ax1.set_xlabel("Chromosome",fontsize=fontsize,family=font_family,labelpad=0, va="center",ha="center")
     ax1.xaxis.set_label_coords(xlabel_coords[0],xlabel_coords[1])
+
+    #ax1.tick_params(axis='x', which='major', pad=xtick_label_pad, labelsize = xtick_label_size)
     
-    ax1.tick_params(axis='x', which='major', pad=xtick_label_pad)
-    
+    for label in ax1.get_xticklabels():
+        label.set_y( xlabel_coords[1] + tick_axes_length )
+    ax1.tick_params(axis='x', which='major', pad=xtick_label_pad, labelsize = xtick_label_size)
+    plt.setp(ax1.get_xticklabels(),  ha='center',va="center")
+
     ax1.set_ylabel("$\mathregular{-log_{10}(P)}$",fontsize=fontsize,family=font_family)
     ax5.set_ylabel("$\mathregular{-log_{10}(P)}$",fontsize=fontsize,family=font_family)
     

gwaslab/viz_plot_mqqplot.py

@@ -1178,22 +1178,33 @@ def _add_pad_to_x_axis(ax1, xpad, xpadl, xpadr, sumstats, pos, chrpad, xtight, l
     
     else:
         chrpad_to_remove = sumstats[pos].max()*chrpad
+        
         if ax1 is not None:
-            xmin, xmax = ax1.get_xlim() 
+            #xmin, xmax = ax1.get_xlim() 
+            xmax = sumstats["i"].max()
+            xmin=  sumstats["i"].min()
+            
+            #length = xmax - xmin
             length = xmax - xmin
             
             if xpad is not None:
                 log.write(" -Adjusting X padding on both side: {}".format(xpad), verbose=verbose)
                 pad = xpad* length #sumstats["i"].max()
                 ax1.set_xlim([xmin - pad + chrpad_to_remove, xmax + pad - chrpad_to_remove])
+
             if xpad is None and xpadl is not None:
                 log.write(" -Adjusting X padding on left side: {}".format(xpadl), verbose=verbose)
-                xmin, xmax = ax1.get_xlim() 
+                
+                xmax =  ax1.get_xlim()[1] 
+
                 pad = xpadl*length # sumstats["i"].max()
                 ax1.set_xlim([xmin - pad + chrpad_to_remove ,xmax])
+            
             if xpad is None and xpadr is not None:
                 log.write(" -Adjusting X padding on right side: {}".format(xpadr), verbose=verbose)
-                xmin, xmax = ax1.get_xlim() 
+                
+                xmin = ax1.get_xlim()[0] 
+
                 pad = xpadr*length # sumstats["i"].max()
                 ax1.set_xlim([xmin, xmax + pad - chrpad_to_remove])
 

{gwaslab-3.6.2.dist-info → gwaslab-3.6.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gwaslab
-Version: 3.6.2
+Version: 3.6.4
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <gwaslab@gmail.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/
@@ -26,6 +26,7 @@ Requires-Dist: pyensembl==2.2.3
 Requires-Dist: gtfparse==1.3.0
 Requires-Dist: h5py>=3.10.0
 Requires-Dist: pyarrow
+Requires-Dist: polars>=1.27.0
 Dynamic: license-file
 
 # GWASLab

{gwaslab-3.6.2.dist-info → gwaslab-3.6.4.dist-info}/RECORD

@@ -2,12 +2,12 @@ gwaslab/__init__.py,sha256=Lpa15i924wRu1jZ9_B2Tz5724FQkdb9O5Wg3tQPWR68,2683
 gwaslab/bd_common_data.py,sha256=jqkdKib-tc8RKyYMPy8z2KF3YwCWIr-6STMCi4xR7KY,13899
 gwaslab/bd_config.py,sha256=TP-r-DPhJD3XnRYZbw9bQHXaDIkiRgK8bG9HCt-UaLc,580
 gwaslab/bd_download.py,sha256=cDDk2C5IvjeAzvPvVYGTkI4Ss33DUtEDjGo8eAbQRvY,15663
-gwaslab/bd_get_hapmap3.py,sha256=qWTvIRZsd7F3nT9sN2NSXUsxZJRf5k4HLgJ6kN0qaUc,4107
+gwaslab/bd_get_hapmap3.py,sha256=1y_mt3Sj-D6Fm4NvVAYJKfE-BPaKPI-8wX93OkmxYbE,5402
 gwaslab/cache_manager.py,sha256=HOTnSkCOyGEPLRl90WT8D_6pAdI8d8AzenMIDGuCeWc,28113
 gwaslab/g_Log.py,sha256=C3Zv-_6c3C9ms8bgQ-ytplz22sjk7euqXYkWr9zNeAs,1573
 gwaslab/g_Phenotypes.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-gwaslab/g_Sumstats.py,sha256=SrF5E46wc_CbVz-GgU3VX-vN9EZeJuVRra4kH3NRI8s,43776
-gwaslab/g_SumstatsMulti.py,sha256=2y8brnKAs5I9tUexQJiFwOsFLC4MyACn7IybvvXsRDU,13828
+gwaslab/g_Sumstats.py,sha256=BKHF4YWPXZBW9XXU9KKkvkhInXZYducvDdC8q7kstbI,43930
+gwaslab/g_SumstatsMulti.py,sha256=kYlbmvq3_PLnr4Car6DV5RIItmvCvznfUAjlYguu-2o,14016
 gwaslab/g_SumstatsPair.py,sha256=PmVPGU32degDuU5RDG9igyT1sFAbEDh6alrRulUphyk,13181
 gwaslab/g_SumstatsSet.py,sha256=AiTISWPfmu8NTGa8j9Yuts8DNw1pEUENYyPoS0HXp5I,29866
 gwaslab/g_SumstatsT.py,sha256=u_DighLMnMxwTLnqm-B58pA0G6WXRj6pudPyKMVKjSU,2133
@@ -18,13 +18,13 @@ gwaslab/g_meta.py,sha256=pXtSsQfFPv2UJmOxKMmLtcsp_Ku_H73YP7PnlfMd0sg,6472
 gwaslab/g_meta_update.py,sha256=dWgz4kcq9bvsXycCjfhoYtcJHlEP4HRdYoQ8KFqMyTQ,2490
 gwaslab/g_vchange_status.py,sha256=w3zsYYOcCaI3PTeboonvkQjudzUAfVIgATzRdiPViZs,1939
 gwaslab/g_vchange_status_polars.py,sha256=kxyGQCur0ibVFBCyZghA-XNf_kLDXKK-l7VC-Om2IdA,1839
-gwaslab/g_version.py,sha256=3oGant2IqVAE1geaqXJxCyH2hrdFPJZWH1T5WpDkVdc,1889
+gwaslab/g_version.py,sha256=YdDG4Pd2xl9SjNiP8yq5slF3aQ4KCaQj83-AWW_edBk,1964
 gwaslab/hm_casting.py,sha256=xoq1E4Tp5VC4aLWfq9-_AfiQzb1WZAHrnZG33W4sCOE,14178
-gwaslab/hm_casting_polars.py,sha256=_3ZeMg3mRsa7vHpxObFErOledpaydhC_PxStcwlYO2c,8371
+gwaslab/hm_casting_polars.py,sha256=CAPfCucj-ARtwP80EuHc9q2gLdj8TL_XQ7_sEC9LxA0,11440
 gwaslab/hm_harmonize_sumstats.py,sha256=2MeWnWmMHpLWjUMTVqExpg_3mE1VIaBsh6Mz3ffJCMc,84761
 gwaslab/hm_rsid_to_chrpos.py,sha256=ODWREO0jPN0RAfNzL5fRzSRANfhiksOvUVPuEsFZQqA,6552
 gwaslab/io_load_ld.py,sha256=pJemMmHn5GrEbIloqGPX3tBStjkA6DVrAoerGOUBbqc,21827
-gwaslab/io_preformat_input.py,sha256=EGSEqoeEtVnUFYV24EeXLkrYjjGZKMxDdpfAmCjp8G8,25661
+gwaslab/io_preformat_input.py,sha256=jIacIVymCfHVBlonUCFRmUJobLcxMWn2w_vWqXjESPg,26538
 gwaslab/io_preformat_input_polars.py,sha256=HGdi6rXPQnYjTW8fMUds-uF6Lt8uElL3Er_Afv3OjTc,24767
 gwaslab/io_process_args.py,sha256=TIv0DyaVEcHnLBGdOD52GBZiO6nAQJycAmmVdEYPmRE,1918
 gwaslab/io_read_ldsc.py,sha256=9tV4AfQZmCAyiNSR9uALmXWOytWhBdT0pfMAY5Mx_QQ,13407
@@ -44,7 +44,7 @@ gwaslab/prscs_parse_genet.py,sha256=N7UmyvdO--yTKvbjMNQCVucRIs4PbRGf93maGkh8PP4,
 gwaslab/qc_build.py,sha256=oeJqV-zbn-9FZ7SsmT4DPuYTt_IIeXw0XtmEkSzN5qA,2145
 gwaslab/qc_check_datatype.py,sha256=XuL4sg-OkdIiGaHg98JHZems0Uo-OyF5DlUTyE9gUd4,4580
 gwaslab/qc_check_datatype_polars.py,sha256=rJ1dIruxclHi37zd5xpS1ml30-200NI-FoEAVTe9_Eo,4218
-gwaslab/qc_fix_sumstats.py,sha256=dZcjMWFmRrH9gRFbt3YVS1t4xuR7zlMvMvbn400cF4s,98438
+gwaslab/qc_fix_sumstats.py,sha256=qjIJxBr_6HpKQLkKRPrbLZEEOr5DzP8Go82AOxgDadY,98448
 gwaslab/qc_fix_sumstats_polars.py,sha256=5DY2PkWiZdfY-k8jXW69YnYPUmJiAw8uaolG8Oztr5g,9600
 gwaslab/run_script.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gwaslab/util_abf_finemapping.py,sha256=LRcopjtkT-iXtKPAJIzR4qjPdhD7nrS_BGit4EW89FM,3054
@@ -60,14 +60,14 @@ gwaslab/util_ex_process_h5.py,sha256=ynFvo3zxgvOxWYL565v2IQf8P6iEuq7UlKQ_ULxrd6Y
 gwaslab/util_ex_process_ref.py,sha256=GQ0ZEWLxGpHLdBs3tqnAqKn3Pqx1A1YvNbYrBLBvXeg,17126
 gwaslab/util_ex_run_2samplemr.py,sha256=eIw-bHlYpgqgRnAhWIWNTbMplVbH4Dbj9kDyDBtzkT0,10423
 gwaslab/util_ex_run_ccgwas.py,sha256=8uOAaJ44zAhlxj_A2vth5AzRLk2_BYlYBXEccFbB6T0,5959
-gwaslab/util_ex_run_clumping.py,sha256=yvUhKi83KhXfE5yPy2i47B58BMACZ_r5gt3-uN8Znbo,7908
+gwaslab/util_ex_run_clumping.py,sha256=7JdHP2XmJS1b0yJObxXDKvEXuZ6wezHvvUMFEzVcswY,8081
 gwaslab/util_ex_run_coloc.py,sha256=evLJ_ct_RV7UKLnYGJ_VfT6MeuGeYDtj8M8xqqk-X-U,6297
 gwaslab/util_ex_run_hyprcoloc.py,sha256=sElGLXsBKx3PpDTOET7ulvjBShtDOJ3OyDuOp3Yl_y0,4722
 gwaslab/util_ex_run_magma.py,sha256=40Mr9IIkUXTHa3bBfROORr-5iVre-W8SeoZhHjh7aTo,2436
 gwaslab/util_ex_run_mesusie.py,sha256=g0x7-zwoN9fbHYJQAue3vAMIeFu3x8H9KjgiYI1u9Mk,5532
 gwaslab/util_ex_run_mtag.py,sha256=TxfDwjbGy5C1SiTVDMNopcSjl9p5mgCjJVBfeLq__bg,2952
 gwaslab/util_ex_run_prscs.py,sha256=_BOUqX3fgALdGx-xDyWlREXCafZZddhcK8ufoESUaac,3396
-gwaslab/util_ex_run_scdrs.py,sha256=_IfWztG8fRYuqClNKiMIGwutDi24d2GtYhjeF6V2jdE,3022
+gwaslab/util_ex_run_scdrs.py,sha256=cs1-6QMB2uixa7j97vN7xe3NQ8aQwZ1QAmfj_cc-Kec,3024
 gwaslab/util_ex_run_susie.py,sha256=YbHishvu7rT83JSHw_Tb_FKzfkZ296YpvNhBesveYR4,6332
 gwaslab/util_in_calculate_gc.py,sha256=MWOXVzJv7SZx4i2_ncRiqsiEOADc7EfghaUzgGy4jaE,2219
 gwaslab/util_in_calculate_power.py,sha256=JfHJFg3tNF0f4NHgWlzVW2mSxCiP07mAHIyEfVfxTak,10360
@@ -85,17 +85,17 @@ gwaslab/util_in_snphwe.py,sha256=-KpIDx6vn_nah6H55IkV2OyjXQVXV13XyBL069WE1wM,175
 gwaslab/viz_aux_annotate_plot.py,sha256=IyGhqFQofrVW9A8UxJ5oNVH96czQ-2_i4s1lVvf2hww,25696
 gwaslab/viz_aux_chromatin.py,sha256=aWZaXOSvGyZY7wQcoFDaqHRYCSHZbi_K4Q70HruN9ts,4125
 gwaslab/viz_aux_property.py,sha256=UIaivghnLXYpTwkKnXRK0F28Jbn9L6OaICk3K73WZaU,33
-gwaslab/viz_aux_quickfix.py,sha256=cGX5i3WBmvKIiqck8V00caDg-pvKOO709Ux3DBXsUrM,18693
+gwaslab/viz_aux_quickfix.py,sha256=v_ZYM2KO39qvk38t0SwopVG3yHjW5MB9DzHX6kvdh1k,18692
 gwaslab/viz_aux_reposition_text.py,sha256=iRIP-Rkltlei068HekJcVubiqPrunBqvAoSQ1eHk04M,4304
 gwaslab/viz_aux_save_figure.py,sha256=HwIRDMYpeXfkBgb7mqzLN7OVPMz163U-ZVOlQJABzeg,2811
 gwaslab/viz_plot_compare_af.py,sha256=qtXW45-Sq_ugK8ZfqBYMpmf58SKi3lB3YyHnzn_akcE,5344
-gwaslab/viz_plot_compare_effect.py,sha256=-v9S8kk06eqBSwny2l0hasqf2z7-rKieGUKnghqyhM0,69370
+gwaslab/viz_plot_compare_effect.py,sha256=QC3TsxKLk1bpiNNO0xh5fQRVcTcETmG3Zr5PXSUX3QY,69480
 gwaslab/viz_plot_credible_sets.py,sha256=RC5ZsuW8-0w0S1h4c--PZrYwfHfgZ3clM-9JcoMLJWk,5841
 gwaslab/viz_plot_effect.py,sha256=qbM6c1IB2HlUlMNgFZlJ5G8ODQJ8-oSWD8t0Q8DDuz8,10653
 gwaslab/viz_plot_forestplot.py,sha256=xgOnefh737CgdQxu5naVyRNBX1NQXPFKzf51fbh6afs,6771
 gwaslab/viz_plot_miamiplot.py,sha256=rCFEp7VNuVqeBBG3WRkmFAtFklbF79BvIQQYiSY70VY,31238
-gwaslab/viz_plot_miamiplot2.py,sha256=eVS2v1YFEO0xZ-2zs_DCCvEx-Hqbt0wQkbgCHiRfqeE,16247
-gwaslab/viz_plot_mqqplot.py,sha256=mlcsA_wEnv7N9mP6CVL15WXWHJ0-8TwqHuznJXgPwCE,71103
+gwaslab/viz_plot_miamiplot2.py,sha256=5O4Ho8_FsSkBcunFVlbAkYuLAMxNqE6swLVtmvkoR6c,17495
+gwaslab/viz_plot_mqqplot.py,sha256=W0rKC8I0KrUvv5zakiTcPTEF_ttrUucrwoFS5MHLIyQ,71283
 gwaslab/viz_plot_phe_heatmap.py,sha256=qoXVeFTIm-n8IinNbDdPFVBSz2yGCGK6QzTstXv6aj4,9532
 gwaslab/viz_plot_qqplot.py,sha256=cB4vRlFv69zWY9NMLfSkfAbirYp3_EEW2kQiBTEMDoc,7483
 gwaslab/viz_plot_regional2.py,sha256=BoL1V56ww9B2_vFkABgln_f6OrzobiFjUISI5X6XXMM,43146
@@ -112,9 +112,9 @@ gwaslab/data/hapmap3_SNPs/hapmap3_db150_hg19.snplist.gz,sha256=qD9RsC5S2h6l-OdpW
 gwaslab/data/hapmap3_SNPs/hapmap3_db151_hg38.snplist.gz,sha256=Y8ZT2FIAhbhlgCJdE9qQVAiwnV_fcsPt72usBa7RSBM,10225828
 gwaslab/data/high_ld/high_ld_hla_hg19.bed.gz,sha256=R7IkssKu0L4WwkU9SrS84xCMdrkkKL0gnTNO_OKbG0Y,219
 gwaslab/data/high_ld/high_ld_hla_hg38.bed.gz,sha256=76CIU0pibDJ72Y6UY-TbIKE9gEPwTELAaIbCXyjm80Q,470
-gwaslab-3.6.2.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-gwaslab-3.6.2.dist-info/licenses/LICENSE_before_v3.4.39,sha256=GhLOU_1UDEKeOacYhsRN_m9u-eIuVTazSndZPeNcTZA,1066
-gwaslab-3.6.2.dist-info/METADATA,sha256=V6dIeyLHIwXppMqbyY8JAZtRK4ra4QGgn2ATIn6AydY,7043
-gwaslab-3.6.2.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
-gwaslab-3.6.2.dist-info/top_level.txt,sha256=PyY6hWtrALpv2MAN3kjkIAzJNmmBTH5a2risz9KwH08,8
-gwaslab-3.6.2.dist-info/RECORD,,
+gwaslab-3.6.4.dist-info/licenses/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+gwaslab-3.6.4.dist-info/licenses/LICENSE_before_v3.4.39,sha256=GhLOU_1UDEKeOacYhsRN_m9u-eIuVTazSndZPeNcTZA,1066
+gwaslab-3.6.4.dist-info/METADATA,sha256=0-B3gOWawfkr19AO_m-H8S-Mgtxe5o_yvCAGCJx59_Y,7073
+gwaslab-3.6.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+gwaslab-3.6.4.dist-info/top_level.txt,sha256=PyY6hWtrALpv2MAN3kjkIAzJNmmBTH5a2risz9KwH08,8
+gwaslab-3.6.4.dist-info/RECORD,,

{gwaslab-3.6.2.dist-info → gwaslab-3.6.4.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.3.1)
+Generator: setuptools (80.9.0)
 Root-Is-Purelib: true
 Tag: py3-none-any