gwaslab 3.6.9__tar.gz → 3.6.10__tar.gz

{gwaslab-3.6.9/src/gwaslab.egg-info → gwaslab-3.6.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gwaslab
-Version: 3.6.9
+Version: 3.6.10
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <gwaslab@gmail.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/

{gwaslab-3.6.9 → gwaslab-3.6.10}/pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "gwaslab"
-version = "3.6.9"
+version = "3.6.10"
 authors = [
   { name="Yunye", email="gwaslab@gmail.com" },
 ]

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/g_Log.py

@@ -31,4 +31,16 @@ class Log():
             self.log_text = self.log_text + str(time.strftime('%Y/%m/%d %H:%M:%S')) + " " + " ".join(map(str,message)) + end
         else:
             if verbose: print(*message,end=end)
-            self.log_text = self.log_text + " ".join(map(str,message)) + end
+            self.log_text = self.log_text + " ".join(map(str,message)) + end
+
+    def get_log_for_last_operation(self):
+        last_log = []
+        rows = self.log_text.strip().split("\n")
+
+        # Iterate backwards to find the last operation
+        for line in reversed(rows):
+            last_log.append(line)
+            if "Start to" in line:
+                break
+
+        return "".join(reversed(last_log))

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/g_Sumstats.py

@@ -112,6 +112,7 @@ from gwaslab.io.io_to_pickle import _reload
 
 #20220309
 class Sumstats():
+    
     def __init__(self,
              sumstats=None,
              fmt=None,
@@ -274,6 +275,10 @@ class Sumstats():
         self.id = id(self)
         self.tmp_path = _path(pid=self.id, log = self.log, verbose=verbose)
 
+    def __getitem__(self, index):
+        return self.data[index]
+    def __len__(self):
+        return len(self.data)
 #### healper #################################################################################
     def update_meta(self, **kwargs):
         self.meta = _update_meta(self.meta, self.data,log = self.log, **kwargs)
@@ -695,7 +700,8 @@ class Sumstats():
         return fig, outliers
     
     def infer_ancestry(self, **kwargs):
-        self.meta["gwaslab"]["inferred_ancestry"] = _infer_ancestry(self.data, **kwargs)
+        self.meta["gwaslab"]["inferred_ancestry"] = _infer_ancestry(self.data, log=self.log,
+                                                                    **kwargs)
 
     def plot_gwheatmap(self, **kwargs):
         fig = _gwheatmap(self.data, **kwargs)

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/g_version.py

@@ -17,8 +17,8 @@ def _get_version():
 def gwaslab_info():
     # version meta information
     dic={
-   "version":"3.6.9",
-   "release_date":"20251012"
+   "version":"3.6.10",
+   "release_date":"20251101"
     }
     return dic   
 

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/util/util_in_calculate_power.py

@@ -27,17 +27,19 @@ def get_power(
         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)
+            log.write("  -Risk allele RR:{}".format(genotype_rr), verbose=verbose)
         elif genotype_or is not None:
-            log.write("  -Risk allele OR:{:.3f}".format(genotype_or), verbose=verbose)
+            log.write("  -Risk allele OR:{}".format(genotype_or), verbose=verbose)
         elif beta is not None:
-            log.write("  -Risk allele beta:{:.3f}".format(beta), verbose=verbose)
+            log.write("  -Risk allele beta:{}".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)
+            log.write("  -Risk allele RR:{}".format(genotype_rr), verbose=verbose)
+
+        log.write("  -Disease prevalence:{}".format(prevalence), verbose=verbose)
+        log.write("  -Risk allele frequency: {}".format(daf), verbose=verbose)
+        log.write("  -Significance level: {}".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)
@@ -68,9 +70,9 @@ def get_power(
         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)
         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)
+        log.write(" - Individuals with AA genotype: {}".format(aap), verbose=verbose)
+        log.write(" - Individuals with AB genotype: {}".format(abp), verbose=verbose)
+        log.write(" - Individuals with BB genotype: {}".format(bbp), verbose=verbose)
         
         pcase= (aap * aaf + abp * abf*0.5) / prevalence
         pcontrol=((1-aap )* aaf + (1-abp )* abf*0.5) / (1 - prevalence)
@@ -78,16 +80,19 @@ def get_power(
         vcase = pcase *(1-pcase)
         vcontrol =pcontrol *(1-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)
+        log.write(" - In cases: {}".format(pcase), verbose=verbose)
+        log.write(" - In controls: {}".format(pcontrol), verbose=verbose)
 
         num= (pcase - pcontrol)
-        den= np.sqrt( (vcase/ncase +  vcontrol/ncontrol)*0.5 )
+        for_sqrt = (vcase/ncase +  vcontrol/ncontrol)*0.5
+        if np.iterable(for_sqrt):
+            for_sqrt[for_sqrt < 0] = np.nan
+        den= np.sqrt( for_sqrt )
         u = num / den
 
         c = ss.norm.isf(sig_level/2)
         power = 1 - ss.norm.cdf(c-u) + ss.norm.cdf(-c-u)
-        log.write("Expected power: {:.3f}".format(power), verbose=verbose)
+        log.write("Expected power: {}".format(power), verbose=verbose)
 
     elif mode=="q":
         if beta is None:

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/viz/viz_plot_associations.py

@@ -33,8 +33,8 @@ def _plot_associations(associations,
     associations = associations.dropna(subset=[values])
     log.write(f" -Keeping associations without NA in {values} :{len(associations)} ", verbose=verbose)
 
-    log.write(f" -Total number of unique variants for plotting :{associations["rsID"].nunique()} ", verbose=verbose)
-    log.write(f" -Total number of unique traits for plotting:{associations["GWASCATALOG_TRAIT"].nunique()} ", verbose=verbose)
+    log.write(f" -Total number of unique variants for plotting :{associations['rsID'].nunique()} ", verbose=verbose)
+    log.write(f" -Total number of unique traits for plotting:{associations['GWASCATALOG_TRAIT'].nunique()} ", verbose=verbose)
 
     matrix_beta = associations.pivot_table(index=['rsID','gene.geneName'], 
                                     columns='GWASCATALOG_TRAIT', 

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/viz/viz_plot_mqqplot.py

@@ -296,7 +296,7 @@ def mqqplot(insumstats,
             region_ref.append(region_ref_second)
     region_ref_index_dic = {value: index for index,value in enumerate(region_ref)}
         
-    taf = _update_args(taf, [track_n,track_n_offset,track_fontsize_ratio,track_exon_ratio,track_text_offset])
+    taf = _update_arg(taf, [track_n,track_n_offset,track_fontsize_ratio,track_exon_ratio,track_text_offset])
     region_marker_shapes = _update_arg(region_marker_shapes, ['o', '^','s','D','*','P','X','h','8'])
     region_grid_line = _update_args(region_grid_line,  {"linewidth": 2,"linestyle":"--"})
     region_lead_grid_line = _update_args(region_lead_grid_line, {"alpha":0.5,"linewidth" : 2,"linestyle":"--","color":"#FF0000"})

{gwaslab-3.6.9 → gwaslab-3.6.10}/src/gwaslab/viz/viz_plot_trumpetplot.py

@@ -170,10 +170,11 @@ def plottrumpet(mysumstats,
             cols_to_use.append(chrom) if chrom not in cols_to_use else cols_to_use
 
     if size != "ABS_BETA":
-        if size not in cols_to_use:
+        if size is not None and size not in cols_to_use:
             cols_to_use.append(size)
     if "hue" in scatter_args.keys():
-        cols_to_use.append(scatter_args["hue"]) 
+        if scatter_args["hue"] not in cols_to_use:
+            cols_to_use.append(scatter_args["hue"]) 
     #filter by p #################################################################################################################
     if p in mysumstats.columns:
         sumstats = mysumstats.loc[mysumstats[p]< p_level,cols_to_use ].copy()
@@ -287,7 +288,10 @@ def plottrumpet(mysumstats,
     sumstats["ABS_BETA"] = sumstats[beta].abs()
 
     ##################################################################################################
-    size_norm = (sumstats[size].min(), sumstats[size].max())
+    if size is None:
+        size_norm = None
+    else:
+        size_norm = (sumstats[size].min(), sumstats[size].max())
     ## if highlight  ##################################################################################################
 
     log.write(" -Creating scatter plot...", verbose=verbose)
@@ -389,15 +393,30 @@ def plottrumpet(mysumstats,
     
     #second_legend = ax.legend(title="Power", loc="upper right",fontsize =fontsize,title_fontsize=fontsize)
     log.write(" -Creating legends...")
+    # curve, size,  hue
     h,l = ax.get_legend_handles_labels()
+    
     if len(ts)>0:
+        # power curves
         l1 = ax.legend(h[:int(len(ts))],l[:int(len(ts))], title="Power", loc="upper right",fontsize =fontsize,title_fontsize=fontsize)
         for line in l1.get_lines():
             line.set_linewidth(5.0)
-    if hue is None:
-        l2 = ax.legend(h[int(len(ts)):],l[int(len(ts)):], title=None, loc="lower right",fontsize =fontsize,title_fontsize=fontsize)
+    ## hue
+    if hue is not None or size is not None:
+    #    # sizes
+        if hue is None and size is not None:
+            hue_size_legend_title = size
+        elif hue is not None and size is None:
+            hue_size_legend_title = hue
+        else:
+            hue_size_legend_title = None
+        l2 = ax.legend(h[int(len(ts)):],l[int(len(ts)):], title=hue_size_legend_title, loc="lower right",fontsize =fontsize,title_fontsize=fontsize)
+    
+    ## hue
     if len(ts)>0:
         ax.add_artist(l1)
+    if hue is not None or size is not None:
+        ax.add_artist(l2)
     #first_legend = ax.legend(handles=dots, loc="lower right" ,title=size,fontsize =fontsize,title_fontsize=fontsize)
     #ax.add_artist(first_legend)
     ##################################################################################################
@@ -780,17 +799,31 @@ def plot_power_x(
                 log=Log()):
     
     #Checking columns#################################################################################################################
+    """
+    Create power cure with user-specified x axis
+    q mode:
+        N
+        MAF
+    b mode:
+        N_CASE
+        N_CONTROL
+        PREVALENCE
+    BETA
+    """
+
     log.write("Start to create power plot...", verbose=verbose)
     matplotlib.rc('font', family=font_family)
 
     log.write(" -Settings:", verbose=verbose)
     log.write("  -Mode: {}".format(mode), verbose=verbose)
+    
     if mode == "q" :
         log.write("  -X axis: {}".format(x), verbose=verbose)
         if x!="N":
             log.write("  -N: {}".format(ns), verbose=verbose)
         if x!="MAF":
             log.write("  -MAF: {}".format(mafs), verbose=verbose)
+
     if mode == "b" :
         log.write("  -X axis: {}".format(x), verbose=verbose)
         if x!="N_CASE":
@@ -846,8 +879,11 @@ def plot_power_x(
         prevalence_range = np.linspace(0.01,0.99,n_matrix)
     else:
         prevalence_range = np.linspace(prevalence_range[0],prevalence_range[1],n_matrix)
-    #configure power threshold###################################################################################################
     
+    
+    #configure power threshold###################################################################################################
+    # for hue
+    # which is the variable
     if type(ns) is list:
         var_to_change = ns
         legend_title = "N"
@@ -895,12 +931,13 @@ def plot_power_x(
     ##creating power line############################################################################################
     if mode=="q":
         for i,value in enumerate(var_to_change):
-            
+            # iterate through variables
             n = ns
             beta = betas
             maf = mafs
             sig_level = sig_levels
             
+            # update the variable
             if legend_title == "BETA":
                 beta = value
             elif legend_title == "MAF":
@@ -910,7 +947,7 @@ def plot_power_x(
             elif legend_title == "Significance level":
                 sig_level = value
 
-
+            # update X
             if x == "N":
                 x_values = n_range
                 n = x_values

{gwaslab-3.6.9 → gwaslab-3.6.10/src/gwaslab.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: gwaslab
-Version: 3.6.9
+Version: 3.6.10
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <gwaslab@gmail.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/