gwaslab 3.4.24__tar.gz → 3.4.25__tar.gz

{gwaslab-3.4.24/src/gwaslab.egg-info → gwaslab-3.4.25}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gwaslab
-Version: 3.4.24
+Version: 3.4.25
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <yunye@gwaslab.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/
@@ -8,7 +8,7 @@ Project-URL: Github, https://github.com/Cloufield/gwaslab
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
+Requires-Python: <=3.9,>=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 
@@ -32,7 +32,7 @@ Note: GWASLab is being updated very frequently for now. I will release the first
 ## Install
 
 ```
-pip install gwaslab==3.4.22
+pip install gwaslab==3.4.24
 ```
 
 

{gwaslab-3.4.24 → gwaslab-3.4.25}/README.md

@@ -18,7 +18,7 @@ Note: GWASLab is being updated very frequently for now. I will release the first
 ## Install
 
 ```
-pip install gwaslab==3.4.22
+pip install gwaslab==3.4.24
 ```
 
 

{gwaslab-3.4.24 → gwaslab-3.4.25}/pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "gwaslab"
-version = "3.4.24"
+version = "3.4.25"
 authors = [
   { name="Yunye", email="yunye@gwaslab.com" },
 ]
@@ -18,7 +18,7 @@ readme = "README.md"
 dependencies = [
     "pandas>=1.3,<1.5",
     "numpy>=1.21.2",
-    "matplotlib>=3.5",
+    "matplotlib>=3.5,<=3.7.1",
     "seaborn>=0.11.1",
     "scipy>=1.6.2",
     "pySAM>=0.18.1,<0.20",
@@ -31,7 +31,7 @@ dependencies = [
     "gtfparse==1.3.0"
 ]
 
-requires-python = ">=3.8"
+requires-python = ">=3.8,<=3.9"
 classifiers = [
     "Programming Language :: Python :: 3",
     "License :: OSI Approved :: MIT License",

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/Sumstats.py

@@ -605,6 +605,7 @@ class Sumstats():
               build="19", 
               n=None,
               verbose=True,
+              no_status=False,
               output_log=True,
               to_csvargs=None,
               float_formats=None,
@@ -713,6 +714,7 @@ class Sumstats():
                   suffix=suffix,
                   build=build,
                   verbose=True,
+                  no_status=no_status,
                   log=onetime_log,
                   to_csvargs=to_csvargs,
                   chr_prefix=chr_prefix,

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/compare_effect.py

@@ -728,6 +728,8 @@ def compare_effect(path1,
         if verbose:log.write(" -Beta_se = ", reg[4])
         #if verbose:log.write(" -H0 beta = ", null_beta, ", recalculated p = ", "{:.2e}".format(p))
         if verbose:log.write(" -H0 beta =  0",", default p = ", "{:.2e}".format(reg[3]))
+        if verbose:log.write(" -Peason correlation coefficient =  ", "{:.2f}".format(reg[2]))
+        if verbose:log.write(" -r2 =  ", "{:.2f}".format(reg[2]**2))
         if r_se==True:
             if verbose:log.write(" -R se (jackknife) = {:.2e}".format(r_se_jackknife))
 

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/download.py

@@ -185,6 +185,7 @@ def get_path(name,log=Log(),verbose=True):
 def download_ref(name,
                 directory=None,
                 local_filename=None,
+                overwrite=False,
                 log=Log()):
     '''
     Download the reference file for a given keyword. Url are retrieved from the reference.json file.
@@ -214,6 +215,9 @@ def download_ref(name,
         # if existing in default path
         if search_local(local_path) == True:
             log.write(" -File {} exists.".format(local_path))
+            if overwrite == True:
+                log.write(" -Overwriting the existing file.")
+                download_file(url,local_path)
         else:
             download_file(url,local_path)
 
@@ -234,6 +238,9 @@ def download_ref(name,
                 try:
                     if search_local(local_path+".tbi") == True:
                         log.write(" -File {} exists.".format(local_path+".tbi"))
+                        if overwrite == True:
+                            log.write(" -Overwriting the existing file.")
+                            download_file(tbi_url,local_path+".tbi")
                     else:
                         download_file(tbi_url,local_path+".tbi")
                     #download_file(tbi_url, local_path+".tbi")

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/fixdata.py

@@ -722,7 +722,7 @@ def sanitycheckstats(sumstats,
                      z=(-37.5,37.5),
                      f=(0,float("Inf")),
                      p=(5e-300,1.000001),
-                     mlog10p=(0,float("Inf")),
+                     mlog10p=(0,9999),
                      beta=(-10,10),
                      se=(0,float("Inf")),
                      OR=(0,10),

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/mqqplot.py

@@ -229,7 +229,7 @@ def mqqplot(insumstats,
     if highlight_anno_args is None:
         highlight_anno_args = {}
     if pinpoint is None:
-        pinpoint = list()
+        pinpoint = list()  
     if build is None:
         build = "19"
     if scatter_kwargs is None:
@@ -255,10 +255,25 @@ def mqqplot(insumstats,
     if len(anno_set)>0 and ("m" in mode):
         if verbose: log.write(" -Variants to annotate : "+",".join(anno_set))    
     if len(highlight)>0 and ("m" in mode):
-        if verbose: log.write(" -Loci to highlight : "+",".join(highlight))    
-        if verbose: log.write("  -Highlight_window is set to: ", highlight_windowkb, " kb")  
+        if pd.api.types.is_list_like(highlight[0]):
+            if len(highlight[0]) == len(highlight_color):
+                log.write(" -WARNING: number of locus list does not match number of colors !!!")
+            for i, highlight_set in enumerate(highlight):
+                  if verbose: log.write(" -Set {} loci to highlight ({}) : ".format(i+1, highlight_color[i%len(highlight_color)])+",".join(highlight_set))   
+            if verbose: log.write("  -Highlight_window is set to: ", highlight_windowkb, " kb") 
+        else:
+            if verbose: log.write(" -Loci to highlight ({}): ".format(highlight_color)+",".join(highlight))    
+            if verbose: log.write("  -Highlight_window is set to: ", highlight_windowkb, " kb") 
+    
     if len(pinpoint)>0 :
-        if verbose: log.write(" -Variants to pinpoint : "+",".join(pinpoint))  
+        if pd.api.types.is_list_like(pinpoint[0]):
+            if len(pinpoint[0]) == len(pinpoint_color):
+                log.write(" -WARNING: number of variant list does not match number of colors !!!")
+            for i, pinpoint_set in enumerate(pinpoint):
+                  if verbose: log.write(" -Set {} variants to pinpoint ({}) : ".format(i+1,pinpoint_color[i%len(pinpoint_color)])+",".join(pinpoint_set))      
+        else:
+            if verbose: log.write(" -Variants to pinpoint ({}) : ".format(pinpoint_color)+",".join(pinpoint))   
+    
     if region is not None:
         if verbose: log.write(" -Region to plot : chr"+str(region[0])+":"+str(region[1])+"-"+str(region[2])+".")  
     
@@ -425,8 +440,9 @@ def mqqplot(insumstats,
         eaf_raw = sumstats["MAF"].copy()
         
     if len(highlight)>0 and ("m" in mode):
-        sumstats["HUE"] = sumstats[chrom].astype("string")
-    
+        sumstats["HUE"] = pd.NA
+        sumstats["HUE"] = sumstats["HUE"].astype("Int64")
+
     if verbose: log.write("Finished loading specified columns from the sumstats.")
 
 
@@ -451,15 +467,27 @@ def mqqplot(insumstats,
             
         ## Highlight
     if len(highlight)>0 and ("m" in mode or "r" in mode):
-        to_highlight = sumstats.loc[sumstats[snpid].isin(highlight),:]
-        #assign colors: 0 is hightlight color
-        for i,row in to_highlight.iterrows():
-            target_chr = int(row[chrom])
-            target_pos = int(row[pos])
-            right_chr=sumstats[chrom]==target_chr
-            up_pos=sumstats[pos]>target_pos-highlight_windowkb*1000
-            low_pos=sumstats[pos]<target_pos+highlight_windowkb*1000
-            sumstats.loc[right_chr&up_pos&low_pos,"HUE"]="0"
+        if pd.api.types.is_list_like(highlight[0]):
+            for i, highlight_set in enumerate(highlight):
+                to_highlight = sumstats.loc[sumstats[snpid].isin(highlight_set),:]
+                #assign colors: 0 is hightlight color
+                for index,row in to_highlight.iterrows():
+                    target_chr = int(row[chrom])
+                    target_pos = int(row[pos])
+                    right_chr=sumstats[chrom]==target_chr
+                    up_pos=sumstats[pos]>target_pos-highlight_windowkb*1000
+                    low_pos=sumstats[pos]<target_pos+highlight_windowkb*1000
+                    sumstats.loc[right_chr&up_pos&low_pos,"HUE"]=i
+        else:
+            to_highlight = sumstats.loc[sumstats[snpid].isin(highlight),:]
+            #assign colors: 0 is hightlight color
+            for index,row in to_highlight.iterrows():
+                target_chr = int(row[chrom])
+                target_pos = int(row[pos])
+                right_chr=sumstats[chrom]==target_chr
+                up_pos=sumstats[pos]>target_pos-highlight_windowkb*1000
+                low_pos=sumstats[pos]<target_pos+highlight_windowkb*1000
+                sumstats.loc[right_chr&up_pos&low_pos,"HUE"]=0
 
 # Density #####################################################################################################              
     if "b" in mode and "DENSITY" not in sumstats.columns:
@@ -612,18 +640,33 @@ def mqqplot(insumstats,
                                sizes=marker_size,
                                linewidth=linewidth,
                                zorder=2,ax=ax1,edgecolor=edgecolor, **scatter_kwargs)   
-            
-            if verbose: log.write(" -Highlighting target loci...")
-            sns.scatterplot(data=sumstats.loc[sumstats["HUE"]=="0"], x='i', y='scaled_P',
-                   hue="HUE",
-                   palette={"0":highlight_color},
-                   legend=legend,
-                   style=style,
-                   size="s",
-                   sizes=(marker_size[0]+1,marker_size[1]+1),
-                   linewidth=linewidth,
-                   zorder=3,ax=ax1,edgecolor=edgecolor,**scatter_kwargs)  
-            highlight_i = sumstats.loc[sumstats[snpid].isin(highlight),"i"].values
+            if pd.api.types.is_list_like(highlight[0]):
+                for i, highlight_set in enumerate(highlight):
+                    if verbose: log.write(" -Highlighting set {} target loci...".format(i+1))
+                    print(sumstats["HUE"].dtype)
+                    sns.scatterplot(data=sumstats.loc[sumstats["HUE"]==i], x='i', y='scaled_P',
+                        hue="HUE",
+                        palette={i:highlight_color[i%len(highlight_color)]},
+                        legend=legend,
+                        style=style,
+                        size="s",
+                        sizes=(marker_size[0]+1,marker_size[1]+1),
+                        linewidth=linewidth,
+                        zorder=3+i,ax=ax1,edgecolor=edgecolor,**scatter_kwargs)  
+                highlight_i = sumstats.loc[~sumstats["HUE"].isna(),"i"].values
+            else:
+                if verbose: log.write(" -Highlighting target loci...")
+                sns.scatterplot(data=sumstats.loc[sumstats["HUE"]==0], x='i', y='scaled_P',
+                    hue="HUE",
+                    palette={0:highlight_color},
+                    legend=legend,
+                    style=style,
+                    size="s",
+                    sizes=(marker_size[0]+1,marker_size[1]+1),
+                    linewidth=linewidth,
+                    zorder=3,ax=ax1,edgecolor=edgecolor,**scatter_kwargs)  
+                # for annotate
+                highlight_i = sumstats.loc[sumstats[snpid].isin(highlight),"i"].values
         
         ## if not highlight    
         else:
@@ -673,12 +716,23 @@ def mqqplot(insumstats,
         
         ## if pinpoint variants
         if (len(pinpoint)>0):
-            if sum(sumstats[snpid].isin(pinpoint))>0:
-                to_pinpoint = sumstats.loc[sumstats[snpid].isin(pinpoint),:]
-                if verbose: log.write(" -Pinpointing target vairants...")
-                ax1.scatter(to_pinpoint["i"],to_pinpoint["scaled_P"],color=pinpoint_color,zorder=3,s=marker_size[1]+1)
+            if pd.api.types.is_list_like(pinpoint[0]):
+                for i, pinpoint_set in enumerate(pinpoint):
+                    if sum(sumstats[snpid].isin(pinpoint_set))>0:
+                        to_pinpoint = sumstats.loc[sumstats[snpid].isin(pinpoint_set),:]
+                        if verbose: log.write(" -Pinpointing set {} target vairants...".format(i+1))
+                        ax1.scatter(to_pinpoint["i"],to_pinpoint["scaled_P"],color=pinpoint_color[i%len(pinpoint_color)],zorder=100,s=marker_size[1]+1)
+                    else:
+                        if verbose: log.write(" -Target vairants to pinpoint were not found. Skip pinpointing process...")
             else:
-                if verbose: log.write(" -Target vairants to pinpoint were not found. Skip pinpointing process...")
+                if sum(sumstats[snpid].isin(pinpoint))>0:
+                    to_pinpoint = sumstats.loc[sumstats[snpid].isin(pinpoint),:]
+                    if verbose: log.write(" -Pinpointing target vairants...")
+                    ax1.scatter(to_pinpoint["i"],to_pinpoint["scaled_P"],color=pinpoint_color,zorder=100,s=marker_size[1]+1)
+                else:
+                    if verbose: log.write(" -Target vairants to pinpoint were not found. Skip pinpointing process...")
+            
+
         
         #ax1.set_xticks(chrom_df.astype("float64"))
         #ax1.set_xticklabels(chrom_df.index.astype("Int64").map(xtick_chr_dict),fontsize=fontsize,family=font_family)

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/read_ldsc.py

@@ -105,52 +105,56 @@ def read_popcorn(filelist=[]):
 #h2 mode
 #####################################################################    
     summary = pd.DataFrame(columns = ["Filename", 'sfile1', 'sfile2', 'mode', 'pg', 'pg_se','pg_z','pg_p', 'h1^2', 'h1^2_se','h1^2_z','h1^2_p', 'h2^2', 'h2^2_se','h2^2_z','h2^2_p'])
-    
+
     for index, ldscfile in enumerate(filelist):
         print("Loading file "+str(index+1)+" :" + ldscfile +" ...")
-        
+
         row={}
-        
-        with open(ldscfile,"r") as file:
-            row["Filename"]=ldscfile.split("/")[-1]
-            line=""
-            while not re.compile('^Invoking command').match(line):
-                line = file.readline()
-                if not line: break
-                    
-            
-                    ## first line h2 se
-            objects = re.compile('--sfile1 ([^\s]+) --sfile2 ([^\s]+)[ /n]').findall(line)
-            row["sfile1"]=objects[0][0]
-            row["sfile2"]=objects[0][1]
-            
-            while not re.compile('^Jackknife iter:').match(line):
-                line = file.readline()
-                if not line: break
-            ##next line lambda gc
-            line = file.readline()
-            
-            #objects = re.compile('[a-zA-Z\s\d]+:|[-0-9.]+[e]?[-0-9.]+|NA').findall(file.readline())
-            objects = file.readline().split()
-            row["h1^2"] = objects[1]
-            row["h1^2_se"] = objects[2]
-            row["h1^2_z"] = objects[3]
-            row["h1^2_p"] = objects[4]
-
-            objects = file.readline().split()
-            row["h2^2"] = objects[1]
-            row["h2^2_se"] = objects[2]
-            row["h2^2_z"] = objects[3]
-            row["h2^2_p"] = objects[4]
-
-            objects = file.readline().split()
-            row["mode"] = objects[0]
-            row["pg"] = objects[1]
-            row["pg_se"] = objects[2]
-            row["pg_z"] = objects[3]
-            row["pg_p"] = objects[4]
+        try:
+            with open(ldscfile,"r") as file:
+                row["Filename"]=ldscfile.split("/")[-1]
+                line=""
+                while not re.compile('^Invoking command').match(line):
+                    line = file.readline()
+                    if not line: break
+
+
+                        ## first line h2 se
+                objects = re.compile('--sfile1 ([^\s]+) --sfile2 ([^\s]+)[ /n]').findall(line)
+                row["sfile1"]=objects[0][0]
+                row["sfile2"]=objects[0][1]
+
+                #while not re.compile(r'^Jackknife iter:').match(line):
+                #    line = file.readline()
+                #    print(line)
+                #    if not line: break
+                while not re.compile(r'P \(Z\)').findall(line.strip()):
+                    line = file.readline()
+                    if not line: break
+
+                #objects = re.compile('[a-zA-Z\s\d]+:|[-0-9.]+[e]?[-0-9.]+|NA').findall(file.readline())
+                objects = file.readline().split()
+                row["h1^2"] = objects[1]
+                row["h1^2_se"] = objects[2]
+                row["h1^2_z"] = objects[3]
+                row["h1^2_p"] = objects[4]
+
+                objects = file.readline().split()
+                row["h2^2"] = objects[1]
+                row["h2^2_se"] = objects[2]
+                row["h2^2_z"] = objects[3]
+                row["h2^2_p"] = objects[4]
+
+                objects = file.readline().split()
+                row["mode"] = objects[0]
+                row["pg"] = objects[1]
+                row["pg_se"] = objects[2]
+                row["pg_z"] = objects[3]
+                row["pg_p"] = objects[4]
+        except:
+            continue
 
         #summary = summary.append(row,ignore_index=True)
         row = pd.DataFrame([row], columns = summary.columns)
         summary = pd.concat([summary, row], ignore_index=True)
-    return summary   
+    return summary

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/to_formats.py

@@ -36,6 +36,7 @@ def tofmt(sumstats,
           bgzip=False,
           tabix=False,
           verbose=True,
+          no_status=False,
           log=Log(),
           to_csvargs=None):
     
@@ -50,7 +51,7 @@ def tofmt(sumstats,
     if verbose: log.write(" - Start outputting sumstats in "+fmt+" format...")
     
     if "CHR" in sumstats.columns:
-        if xymt_number is False and sumstats["CHR"].dtype in ["Int64","int"]:
+        if xymt_number is False and pd.api.types.is_integer_dtype(sumstats["CHR"]):
             sumstats["CHR"]= sumstats["CHR"].map(get_number_to_chr(xymt=xymt,prefix=chr_prefix))
         elif chr_prefix is not None:
             sumstats["CHR"]= chr_prefix + sumstats["CHR"].astype("string")
@@ -364,7 +365,11 @@ def tofmt(sumstats,
                 ouput_cols.append(i)  
         # + additional cols
         ouput_cols = ouput_cols + cols
-        
+        try:
+            if no_status == True:
+                ouput_cols.remove("STATUS")
+        except:
+            pass
         sumstats = sumstats.loc[:,ouput_cols]
         sumstats = sumstats.rename(columns=rename_dictionary) 
         

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/trumpetplot.py

@@ -46,13 +46,15 @@ def plottrumpet(mysumstats,
                 yscale_factor=1,
                 cmap="cool",
                 ylim=None,
+                xlim=None,
                 markercolor="#597FBD",
                 fontsize=15,
                 font_family="Arial",
-                size= "BETA",
+                size= "ABS_BETA",
                 sizes=None,
                 save=False,
                 saveargs=None,
+                figargs=None,
                 build="99",
                 anno_set=None,
                 anno_alias=None,
@@ -94,7 +96,8 @@ def plottrumpet(mysumstats,
         else:
             xticks = [0,0.01,0.05,0.1,0.2,0.5]
             xticklabels = xticks            
-
+    if figargs is None:
+        figargs={"figsize":(10,15)}
     #Checking columns#################################################################################################################
     if verbose: log.write("Start to create trumpet plot...")
     
@@ -201,7 +204,7 @@ def plottrumpet(mysumstats,
     output_hex_colors
 
     ##################################################################################################
-    fig, ax = plt.subplots(figsize=(10,10))
+    fig, ax = plt.subplots(**figargs)
     
     ##creating power line############################################################################################
     if mode=="q":
@@ -284,6 +287,9 @@ def plottrumpet(mysumstats,
         rotation=90    
         ax.set_xticks(xticks,xticklabels,fontsize=fontsize,rotation=rotation)
         ax.set_xlim(-0.02,0.52)
+        
+    if xlim is not None:
+        ax.set_xlim(xlim)
     
     if ylim is not None:
         ax.set_ylim(ylim)

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab/version.py

@@ -8,7 +8,7 @@ def _show_version(log=Log()):
 def gwaslab_info():
     # for output header
     dic={
-   "version":"3.4.24",
-   "release_date":"20230821"
+   "version":"3.4.25",
+   "release_date":"20230905"
     }
     return dic   

{gwaslab-3.4.24 → gwaslab-3.4.25/src/gwaslab.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gwaslab
-Version: 3.4.24
+Version: 3.4.25
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <yunye@gwaslab.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/
@@ -8,7 +8,7 @@ Project-URL: Github, https://github.com/Cloufield/gwaslab
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
+Requires-Python: <=3.9,>=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 
@@ -32,7 +32,7 @@ Note: GWASLab is being updated very frequently for now. I will release the first
 ## Install
 
 ```
-pip install gwaslab==3.4.22
+pip install gwaslab==3.4.24
 ```
 
 

{gwaslab-3.4.24 → gwaslab-3.4.25}/src/gwaslab.egg-info/requires.txt

@@ -1,6 +1,6 @@
 pandas<1.5,>=1.3
 numpy>=1.21.2
-matplotlib>=3.5
+matplotlib<=3.7.1,>=3.5
 seaborn>=0.11.1
 scipy>=1.6.2
 pySAM<0.20,>=0.18.1