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

gwaslab/viz_aux_annotate_plot.py

@@ -58,7 +58,7 @@ def annotate_single(
                 annotation_col="CHR:POS"
         elif anno:
                 annotation_col=anno
-        if verbose: log.write(" -Annotating using column "+annotation_col+"...")
+        log.write(" -Annotating using column "+annotation_col+"...", verbose=verbose)
         
         ## calculate y span
         if region is not None:
@@ -66,7 +66,7 @@ def annotate_single(
         else:
             y_span = sumstats["i"].max()-sumstats["i"].min()
         
-        if verbose: log.write(" -Adjusting text positions with repel_force={}...".format(repel_force))
+        log.write(" -Adjusting text positions with repel_force={}...".format(repel_force), verbose=verbose)
         if anno_style == "expand" :
             to_annotate.loc[:, "ADJUSTED_i"] = adjust_text_position(to_annotate["i"].values.copy(), y_span, repel_force,max_iter=anno_max_iter,log=log,amode=amode,verbose=verbose)
         ##  iterate through variants to be annotated
@@ -214,7 +214,7 @@ def annotate_single(
             anno_count +=1
         #anno_adjust_keyargs = {"arrowprops":dict(arrowstyle='->', color='grey', linewidth=0.1,relpos=(0.5,0.5))}
         if anno_adjust==True:
-            if verbose: log.write(" -Auto-adjusting text positions...")
+            log.write(" -Auto-adjusting text positions...", verbose=verbose)
             adjust_text(texts = anno_to_adjust_list,
                         autoalign=False, 
                         only_move={'points':'x', 'text':'x', 'objects':'x'},
@@ -232,7 +232,7 @@ def annotate_single(
                         )
         
     else:
-        if verbose: log.write(" -Skip annotating")
+        log.write(" -Skip annotating", verbose=verbose)
     
     return ax1
 
@@ -275,7 +275,7 @@ def annotate_pair(
         for index,ax,to_annotate_df,anno_d, anno_alias in [(0,ax1,to_annotate1,anno_d1,anno_alias1),(1,ax5,to_annotate5,anno_d2,anno_alias2)]:
             ###################### annotate() args
             if to_annotate_df.empty is True:
-                if verbose: log.write(" -Skipping annotation...")
+                log.write(" -Skipping annotation...", verbose=verbose)
                 continue
 
             fontweight = "normal"
@@ -313,7 +313,7 @@ def annotate_pair(
                             annotation_col=anno
                         else:
                             annotation_col=anno+"_"+str(index+1)
-                if verbose: log.write(" -Annotating using column "+annotation_col+"...")
+                log.write(" -Annotating using column "+annotation_col+"...", verbose=verbose)
                 
                 ## calculate y span
                 if region is not None:
@@ -453,7 +453,7 @@ def annotate_pair(
                     anno_count +=1
             
             if anno_adjust==True:
-                if verbose: log.write(" -Auto-adjusting text positions for plot {}...".format(index))
+                log.write(" -Auto-adjusting text positions for plot {}...".format(index), verbose=verbose)
                 if index==0:
                     va="bottom"
                     ha='left'
@@ -476,7 +476,7 @@ def annotate_pair(
                             lim =anno_max_iter
                             )
     else:
-        if verbose: log.write(" -Skip annotating")
+        log.write(" -Skip annotating", verbose=verbose)
     return ax1,ax5
 
 
@@ -521,7 +521,7 @@ def annotate_subtype(
                 annotation_col="CHR:POS"
         elif anno:
                 annotation_col=anno
-        if verbose: log.write(" -Annotating using column "+annotation_col+"...")
+        log.write(" -Annotating using column "+annotation_col+"...", verbose=verbose)
         
         ## calculate y span
         if region is not None:
@@ -529,7 +529,7 @@ def annotate_subtype(
         else:
             y_span = sumstats["i"].max()-sumstats["i"].min()
         
-        if verbose: log.write(" -Adjusting text positions with repel_force={}...".format(repel_force))
+        log.write(" -Adjusting text positions with repel_force={}...".format(repel_force), verbose=verbose)
         if anno_style == "expand" :
             to_annotate.loc[:, "ADJUSTED_i"] = adjust_text_position(to_annotate["i"].values.copy(), y_span, repel_force,max_iter=anno_max_iter,log=log,verbose=verbose)
         ##  iterate through variants to be annotated
@@ -656,7 +656,7 @@ def annotate_subtype(
             anno_count +=1
         #anno_adjust_keyargs = {"arrowprops":dict(arrowstyle='->', color='grey', linewidth=0.1,relpos=(0.5,0.5))}
         if anno_adjust==True:
-            if verbose: log.write(" -Auto-adjusting text positions...")
+            log.write(" -Auto-adjusting text positions...", verbose=verbose)
             adjust_text(texts = anno_to_adjust_list,
                         autoalign=False, 
                         only_move={'points':'x', 'text':'x', 'objects':'x'},
@@ -674,7 +674,7 @@ def annotate_subtype(
                         )
         
     else:
-        if verbose: log.write(" -Skip annotating")
+        log.write(" -Skip annotating", verbose=verbose)
     
     return ax1
 

gwaslab/viz_aux_quickfix.py

@@ -31,16 +31,16 @@ def _quick_fix_p_value(sumstats, p="P", mlog10p="MLOG10P", scaled=False,verbose=
     '''
     if scaled==True:
         # if scaled, add scaled P and P col 
-        if verbose:log.write(" -P values are already scaled...")
-        if verbose:log.write(" -Sumstats -log10(P) values are being converted to P...")
+        log.write(" -P values are already scaled...", verbose=verbose)
+        log.write(" -Sumstats -log10(P) values are being converted to P...", verbose=verbose)
         sumstats["scaled_P"] = sumstats[mlog10p].copy()
         sumstats[p]= np.power(10,-sumstats[mlog10p].astype("float64"))
         return sumstats
     # bad p : na and outside (0,1]
     bad_p_value = (sumstats[p].isna()) | (sumstats[p] > 1) | (sumstats[p] <= 0)
-    if verbose:
-        log.write(" -Sanity check after conversion: " + str(sum(bad_p_value)) +
-                  " variants with P value outside of (0,1] will be removed...")
+
+    log.write(" -Sanity check after conversion: " + str(sum(bad_p_value)) +
+                  " variants with P value outside of (0,1] will be removed...", verbose=verbose)
     sumstats = sumstats.loc[~bad_p_value, :]
     return sumstats
 
@@ -51,15 +51,15 @@ def _quick_fix_mlog10p(insumstats,p="P", mlog10p="MLOG10P", scaled=False, log=Lo
     '''
     sumstats = insumstats.copy()
     if scaled != True:
-        if verbose:log.write(" -Sumstats P values are being converted to -log10(P)...")
+        log.write(" -Sumstats P values are being converted to -log10(P)...", verbose=verbose)
         sumstats["scaled_P"] = -np.log10(sumstats[p].astype("float64"))
         
     #with pd.option_context('mode.use_inf_as_na', True):
     #    is_na = sumstats["scaled_P"].isna()
     if_inf_na = np.isinf(sumstats["scaled_P"]) | sumstats["scaled_P"].isna()
-    if verbose:
-        log.write(" -Sanity check: "+str(sum(if_inf_na)) +
-                  " na/inf/-inf variants will be removed...")
+
+    log.write(" -Sanity check: "+str(sum(if_inf_na)) +
+                  " na/inf/-inf variants will be removed...", verbose=verbose)
     sumstats = sumstats.loc[~if_inf_na, :]
     return sumstats
 
@@ -105,7 +105,7 @@ def _get_largenumber(*args,log=Log(), verbose=True):
         else:
             break
         if i == 7:
-            log.write(" -Warning: max POS is too large!")
+            log.warning("Max POS is too large!")
     return large_number
 
 
@@ -276,9 +276,9 @@ def _quick_extract_snp_in_region(sumstats, region, chrom="CHR",pos="POS",log=Log
     region_chr = region[0]
     region_start = region[1]
     region_end = region[2]
-    if verbose:log.write(" -Extract SNPs in region : chr"+str(region_chr)+":"+str(region[1])+"-"+str(region[2])+ "...")
+    log.write(" -Extract SNPs in region : chr"+str(region_chr)+":"+str(region[1])+"-"+str(region[2])+ "...", verbose=verbose)
     is_in_region_snp = (sumstats[chrom]==region_chr) &(sumstats[pos]<region_end) &(sumstats[pos]>region_start)
-    if verbose:log.write(" -Extract SNPs in specified regions: "+str(sum(is_in_region_snp)))
+    log.write(" -Extract SNPs in specified regions: "+str(sum(is_in_region_snp)), verbose=verbose)
     sumstats = sumstats.loc[is_in_region_snp,:]
     return sumstats
 
@@ -289,18 +289,18 @@ def _cut(series, mode,cutfactor,cut,skip, ylabels, cut_log, verbose, lines_to_pl
     maxy = series.max()
     series = series.copy()
     if "b" not in mode:
-        if verbose: log.write(" -Maximum -log10(P) value is "+str(maxy) +" .")
+        log.write(" -Maximum -log10(P) value is "+str(maxy) +" .", verbose=verbose)
     elif "b" in mode:
-        if verbose: log.write(" -Maximum DENSITY value is "+str(maxy) +" .")
+        log.write(" -Maximum DENSITY value is "+str(maxy) +" .", verbose=verbose)
     
     maxticker=int(np.round(series.max(skipna=True)))
     
     if cut:
         # auto mode : determine curline and cut factor
         if cut==True:
-            if verbose: log.write(" -Cut Auto mode is activated...")
+            log.write(" -Cut Auto mode is activated...", verbose=verbose)
             if maxy<30:
-                if verbose: log.write(" - maxy <30 , no need to cut.")
+                log.write(" - maxy <30 , no need to cut.", verbose=verbose)
                 cut=0
             else:
                 cut = 20
@@ -327,9 +327,9 @@ def _cut(series, mode,cutfactor,cut,skip, ylabels, cut_log, verbose, lines_to_pl
             else:
                 # cut linear mode
                 if "b" not in mode:
-                    if verbose: log.write(" -Minus log10(P) values above " + str(cut)+" will be shrunk with a shrinkage factor of " + str(cutfactor)+"...")
+                    log.write(" -Minus log10(P) values above " + str(cut)+" will be shrunk with a shrinkage factor of " + str(cutfactor)+"...", verbose=verbose)
                 else:
-                    if verbose: log.write(" -Minus DENSITY values above " + str(cut)+" will be shrunk with a shrinkage factor of " + str(cutfactor)+"...")
+                    log.write(" -Minus DENSITY values above " + str(cut)+" will be shrunk with a shrinkage factor of " + str(cutfactor)+"...", verbose=verbose)
 
                 maxticker=int(np.round(series.max(skipna=True)))
 

gwaslab/viz_aux_reposition_text.py

@@ -6,10 +6,10 @@ def adjust_text_position(positions, yspan, repel_force=0.01, max_iter=100,amode=
     # check the number of variants to annotate 
     #if repel_force>0:
     #    if 1/(repel_force*2 +0.01) < len(positions):
-    #        if verbose: log.write(" -Too many variants to annotate; maybe it is better to reduce the number of variants")
+    #        log.write(" -Too many variants to annotate; maybe it is better to reduce the number of variants")
     #else:
     if len(positions)>30:
-        if verbose: log.write(" -Too many variants to annotate; maybe it is better to reduce the number of variants")
+        log.write(" -Too many variants to annotate; maybe it is better to reduce the number of variants",verbose=verbose)
 
     # calculate the steps
     if amode=="int":
@@ -47,7 +47,7 @@ def adjust_text_position(positions, yspan, repel_force=0.01, max_iter=100,amode=
                 move_position_from_center_float(positions, index, step)
     
     # when reaching maximum iteration, return anyway
-    if verbose: log.write(" -Reaching maximum iteration: {}; Skipping...".format(max_iter))
+    log.write(" -Reaching maximum iteration: {}; Skipping...".format(max_iter),verbose=verbose)
     if amode=="int":
         return np.floor(pd.to_numeric(positions, errors='coerce')).astype('Int64').copy()
     elif amode=="log":

gwaslab/viz_aux_save_figure.py

@@ -7,18 +7,27 @@ def save_figure(fig, save, keyword, save_args=None, log = Log(), verbose=True):
     log.write("Start to save figure..." ,verbose=verbose)
     if save_args is None:
         save_args = {}
+        
     if save:
         if save==True:
             default_path = get_default_path(keyword)
             fig.savefig(default_path, bbox_inches="tight",**save_args)
             log.write(" -Saved to "+ default_path + " successfully!" ,verbose=verbose)
         else:
-            if os.path.exists(save):
-                fig.savefig(save,bbox_inches="tight",**save_args)
-                log.write(" -Saved to "+ save + " successfully! (overwrite)" ,verbose=verbose)
+            if save[-3:]=="pdf":
+                if os.path.exists(save):
+                    fig.savefig(save, **save_args)
+                    log.write(" -Saved to "+ save + " successfully! (pdf, overwrite)" ,verbose=verbose)
+                else:
+                    fig.savefig(save, **save_args)
+                    log.write(" -Saved to "+ save + " successfully! (pdf)" ,verbose=verbose)
             else:
-                fig.savefig(save,bbox_inches="tight",**save_args)
-                log.write(" -Saved to "+ save + " successfully!" ,verbose=verbose)
+                if os.path.exists(save):
+                    fig.savefig(save,bbox_inches="tight",**save_args)
+                    log.write(" -Saved to "+ save + " successfully! (overwrite)" ,verbose=verbose)
+                else:
+                    fig.savefig(save,bbox_inches="tight",**save_args)
+                    log.write(" -Saved to "+ save + " successfully!" ,verbose=verbose)
     else:
         log.write(" -Skip saving figure!" ,verbose=verbose)
     log.write("Finished saving figure..." ,verbose=verbose)

gwaslab/viz_plot_compare_af.py

@@ -10,6 +10,7 @@ from gwaslab.viz_aux_save_figure import save_figure
 def plotdaf(sumstats,
              eaf="EAF",
              daf="DAF",
+             raf="RAF",
              threshold=0.16,
              xlabel="Alternative Allele Frequency in Reference Population (RAF)",
              ylabel="Effect Allele Frequency in Sumstats (EAF)",
@@ -43,7 +44,7 @@ def plotdaf(sumstats,
     if plt_args is None:
         plt_args={"figsize":(8,4),"dpi":300}
     if histplot_args is None:
-        histplot_args={"log_scale":(False,True)}
+        histplot_args={"log_scale":(False,False)}
     if reg_line_args is None:
         reg_line_args={"color":'#cccccc', "linestyle":'--'}
     if threshold_line_args is None:
@@ -55,8 +56,9 @@ def plotdaf(sumstats,
     if save_args is None:
         save_args =  {}
 
-    if verbose: log.write("Start to plot Reference frequency vs Effect allele frequency plot...")
-    if not ((eaf in sumstats.columns) and (daf in sumstats.columns)):
+    log.write("Start to plot allele frequency comparison plot...", verbose=verbose)
+    
+    if not ((eaf in sumstats.columns) and ((daf in sumstats.columns)) or (raf in sumstats.columns)):
         raise ValueError("EAF and/or DAF columns were not detected.")
     
     if "SNPID" in sumstats.columns:
@@ -70,40 +72,42 @@ def plotdaf(sumstats,
     if "NEA" in sumstats.columns:
         alleles.append("NEA")
     
+    if daf not in sumstats.columns:
+        sumstats[daf] = sumstats[eaf] - sumstats[raf]
 
     sumstats = sumstats.loc[(~sumstats[eaf].isna())&(~sumstats[daf].isna()),[snpid,eaf,daf]+alleles].copy()
     sumstats[daf] = sumstats[daf].astype("float")
     sumstats[eaf] = sumstats[eaf].astype("float")
-    if verbose: log.write(" -Plotting valriants:" + str(len(sumstats)))
-    
-    sumstats["RAF"]=sumstats[eaf] - sumstats[daf]
+    log.write(" -Plotting valriants:" + str(len(sumstats)), verbose=verbose)
+    if raf not in sumstats.columns:
+        sumstats[raf] = sumstats[eaf] - sumstats[daf]
     sns.set_style("ticks")
-    fig, (ax1, ax2) = plt.subplots(1, 2,**plt_args)
-    ax1.scatter(sumstats["RAF"],sumstats[eaf],label="Non-outlier", **scatter_args)
+    fig, [ax1, ax2] = plt.subplots(1, 2,**plt_args)
+    ax1.scatter(sumstats[raf],sumstats[eaf],label="Non-outlier", **scatter_args)
     
     if is_threshold is True:
         is_outliers = sumstats[daf].abs() > threshold 
         if sum(is_outliers)>0:
-            ax1.scatter(sumstats.loc[is_outliers, "RAF"],sumstats.loc[is_outliers, eaf],label="Outlier", **scatter_args_outlier)
+            ax1.scatter(sumstats.loc[is_outliers, raf],sumstats.loc[is_outliers, eaf],label="Outlier", **scatter_args_outlier)
     
     if legend1 ==True:
         ax1.legend()
     
     if is_reg is True:
-        if verbose: log.write(" -Plotting regression line...")
-        reg = ss.linregress(sumstats["RAF"],sumstats[eaf])
-        if verbose:log.write(" -Beta = ", reg[0])
-        if verbose:log.write(" -Intercept = ", reg[1])
-        if verbose:log.write(" -R2 = ", reg[2])
+        log.write(" -Plotting regression line...", verbose=verbose)
+        reg = ss.linregress(sumstats[raf],sumstats[eaf])
+        log.write(" -Beta = ", reg[0], verbose=verbose)
+        log.write(" -Intercept = ", reg[1], verbose=verbose)
+        log.write(" -R2 = ", reg[2], verbose=verbose)
         ax1.axline(xy1=(0,reg[1]),slope=reg[0],zorder=1,**reg_line_args)
         if r2 is True:
             ax1.text(0.98,0.02, "$R^2 = {:.3f}$".format(reg[2]), transform=ax1.transAxes, **r2_args)
 
     if is_threshold is True:
-        if verbose: log.write(" -Threshold : " + str(threshold))
+        log.write(" -Threshold : " + str(threshold), verbose=verbose)
         num = sum(np.abs(sumstats[daf])>threshold )
-        if verbose: log.write(" -Variants with relatively large DAF : ",num )
-        if verbose: log.write(" -Percentage for variants with relatively large DAF : ",num/len(sumstats) )
+        log.write(" -Variants with relatively large DAF : ",num , verbose=verbose)
+        log.write(" -Percentage for variants with relatively large DAF : ",num/len(sumstats) , verbose=verbose)
         ax1.axline(xy1=(0,threshold),slope=1,zorder=1,**threshold_line_args)
         ax1.axline(xy1=(threshold,0),slope=1,zorder=1,**threshold_line_args)
     
@@ -121,23 +125,18 @@ def plotdaf(sumstats,
 
     sumstats["ID"] = sumstats.index
     
-    to_plot = pd.melt(sumstats,id_vars=['ID'], value_vars=['EAF',"RAF"], var_name='Types', value_name='Allele Frequency').dropna()
-    
-    sns.histplot(data=to_plot, x="Allele Frequency", hue="Types", fill=True, ax=ax2, legend = legend2 ,**histplot_args)
-    ax2.set_xlabel("Allele Frequency",**font_args)
+    to_plot = pd.melt(sumstats,id_vars=['ID'], value_vars=[eaf,raf], var_name='Types', value_name='Allele Frequency').dropna()
 
+    sns.histplot(data=to_plot, x="Allele Frequency", 
+                 hue="Types", fill=True, 
+                 ax=ax2, legend = legend2,
+                 **histplot_args)
+ 
+    ax2.set_xlabel("Allele Frequency",**font_args)
 
     plt.tight_layout()
     save_figure(fig, save, keyword="afc",save_args=save_args, log=log, verbose=verbose)
-
-    #if save:
-    #    if verbose: log.write("Saving plot:")
-    #    if save==True:
-    #        fig.savefig("./allele_frequency_comparison.png",bbox_inches="tight",**save_args)
-    #        log.write(" -Saved to "+ "./allele_frequency_comparison.png" + " successfully!" )
-    #    else:
-    #        fig.savefig(save,bbox_inches="tight",**save_args)
-    #        log.write(" -Saved to "+ save + " successfully!" )
     sumstats = sumstats.drop(columns="ID")
+
     return fig, sumstats[is_outliers].copy()