gwaslab 3.6.3__tar.gz → 3.6.5__tar.gz

{gwaslab-3.6.3/src/gwaslab.egg-info → gwaslab-3.6.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gwaslab
-Version: 3.6.3
+Version: 3.6.5
 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-3.6.3 → gwaslab-3.6.5}/pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "gwaslab"
-version = "3.6.3"
+version = "3.6.5"
 authors = [
   { name="Yunye", email="gwaslab@gmail.com" },
 ]
@@ -29,7 +29,8 @@ dependencies = [
     "pyensembl==2.2.3",
     "gtfparse==1.3.0",
     "h5py>=3.10.0",
-    "pyarrow"
+    "pyarrow",
+    "polars>=1.27.0"
 ]
 
 requires-python = ">=3.9,<3.13"

{gwaslab-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/gwaslab/g_Sumstats.py

@@ -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-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/gwaslab/g_version.py

@@ -17,8 +17,8 @@ def _get_version():
 def gwaslab_info():
     # version meta information
     dic={
-   "version":"3.6.3",
-   "release_date":"202505"
+   "version":"3.6.5",
+   "release_date":"20250625"
     }
     return dic   
 

{gwaslab-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/gwaslab/io_load_ld.py

@@ -233,9 +233,7 @@ def process_ld(sumstats,
                 lead_snpid = matched_snpid.group(0).split(":")[1:]
                 lead_snpid[0]= int(lead_snpid[0])
                 lead_series = pd.Series(lead_snpid)
-                
-        print(sumstats)
-        print(sumstats.loc[lead_id, "REFINDEX"])
+
         # if lead pos is available: 
         if sumstats.loc[lead_id, "REFINDEX"] is not None:
             lead_snp_ref_index = sumstats.loc[lead_id, "REFINDEX"]

{gwaslab-3.6.3 → gwaslab-3.6.5}/src/gwaslab/io_preformat_input.py

@@ -59,7 +59,8 @@ def preformat(sumstats,
           trait=None,
           build=None,
           other=None,
-          usekeys=None,
+          exclude=None,
+          include=None,
           chrom_pat=None,
           snpid_pat=None,
           verbose=False,
@@ -71,6 +72,11 @@ def preformat(sumstats,
     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={}
@@ -78,7 +84,7 @@ def preformat(sumstats,
     # 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)
@@ -318,18 +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-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5}/src/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,
@@ -101,9 +102,16 @@ def plot_miami2(
             cols = ["CHR","POS","P"]
 
     if cols1 is None:
-        cols1 = cols.copy()
+        if scaled1 == True:
+            cols1 = ["CHR","POS","MLOG10P"]
+        else:
+            cols1 = ["CHR","POS","P"]
+    
     if cols2 is None:
-        cols2 = cols.copy()
+        if scaled2 == True:
+            cols2 = ["CHR","POS","MLOG10P"]
+        else:
+            cols2 = ["CHR","POS","P"]
 
     if id1 is not None:
         cols1.append(id1)
@@ -142,9 +150,12 @@ 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 +168,10 @@ 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]
@@ -211,8 +213,15 @@ def plot_miami2(
         sumstats2 = merged_sumstats[cols2].copy()
 
     ## rename and quick fix ###########################################################################################################
-    renaming_dict1 = {cols1[0]:"CHR",cols1[1]:"POS",cols1[2]:"P"}
-    renaming_dict2 = {cols2[0]:"CHR",cols2[1]:"POS",cols2[2]:"P"}
+    if scaled1==True:
+        renaming_dict1 = {cols1[0]:"CHR",cols1[1]:"POS",cols1[2]:"MLOG10P"}
+    else:
+        renaming_dict1 = {cols1[0]:"CHR",cols1[1]:"POS",cols1[2]:"P"}
+    if scaled2==True:
+        renaming_dict2 = {cols2[0]:"CHR",cols2[1]:"POS",cols2[2]:"MLOG10P"}
+    else:
+        renaming_dict2 = {cols2[0]:"CHR",cols2[1]:"POS",cols2[2]:"P"}
+    
 
     sumstats1 = sumstats1.rename(columns=renaming_dict1)
     sumstats1 = _quick_fix(sumstats1,chr_dict=chr_dict1, scaled=scaled1, verbose=verbose, log=log)
@@ -242,11 +251,12 @@ 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 +264,30 @@ 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()
 
@@ -273,6 +306,7 @@ def plot_miami2(
                       _chrom_df_for_i = chrom_df,
                       _invert=False,
                       _if_quick_qc=False,
+                      font_family=font_family,
                       **mqq_args1
                      )
     log.write("Finished creating Manhattan plot for sumstats1".format(_get_version()), verbose=verbose)
@@ -292,6 +326,7 @@ def plot_miami2(
                       _chrom_df_for_i = chrom_df,
                        _invert=True,
                       _if_quick_qc=False,
+                      font_family=font_family,
                      **mqq_args2)
     log.write("Finished creating Manhattan plot for sumstats2".format(_get_version()), verbose=verbose)
     
@@ -307,17 +342,23 @@ 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-3.6.3 → gwaslab-3.6.5}/src/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-3.6.3 → gwaslab-3.6.5/src/gwaslab.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gwaslab
-Version: 3.6.3
+Version: 3.6.5
 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-3.6.3 → gwaslab-3.6.5}/src/gwaslab.egg-info/requires.txt

@@ -12,3 +12,4 @@ pyensembl==2.2.3
 gtfparse==1.3.0
 h5py>=3.10.0
 pyarrow
+polars>=1.27.0