gwaslab 3.4.15__py3-none-any.whl → 3.4.16__py3-none-any.whl

gwaslab/__init__.py

@@ -32,4 +32,6 @@ from gwaslab.download import update_record
 from gwaslab.to_pickle import dump_pickle
 from gwaslab.to_pickle import load_pickle
 from gwaslab.config import options
-from gwaslab.version import _show_version as show_version
+from gwaslab.version import _show_version as show_version
+from gwaslab.calculate_power import get_power
+from gwaslab.calculate_power import get_beta

gwaslab/calculate_power.py

@@ -1,47 +1,124 @@
-def get_power(genotype_or=1.3 ,
+import pandas as pd
+import numpy as np
+import scipy.stats as ss
+from gwaslab.Log import Log
+import scipy as sp
+
+def get_power(
+              mode="b",
+              t=0,
+              genotype_or=1.3 ,
+              beta=0.3,
+              eaf=0.1,
+              n=10000,
               scase= 2000,
               scontrol= 15000,
               prevalence= 0.15,
               daf = 0.2,
-              sig_level= 5e-8
+              sig_level= 5e-8,
+              vary=1,
+              log=Log(),
+              verbose=True
              ):
-    print("Input settings:{}".format(daf))
-    print(" -Number of cases:{}".format(scase))
-    print(" -Number of controls:{}".format(scontrol))
-    print(" -Risk allele OR:{:.3f}".format(genotype_or))
-    print(" -Disease prevalence:{:.3f}".format(prevalence))
-    print(" -Risk allele frequency: {:.3f}".format(daf))
-    print(" -Significance level: {:.3e}".format(sig_level))
-    # 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)
-    bbf = (1- daf)**2
-
-    # additive
-    x = [ 2*genotype_or-1, genotype_or, 1 ] 
-
-    aap= x[0] * prevalence / (x[0]*aaf + x[1]*abf + x[2]*bbf)
-    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)
-    print("Probability of disease :")
-    print(" - Individuals with AA genotype: {:.3f}".format(aap))
-    print(" - Individuals with AB genotype: {:.3f}".format(abp))
-    print(" - Individuals with BB genotype: {:.3f}".format(bbp))
-    
-    pcase= (aap * aaf + abp * abf*0.5) / prevalence
-    pcontrol=((1-aap )* aaf + (1-abp )* abf*0.5) / (1 - prevalence)
-
-    vcase = pcase *(1-pcase)
-    vcontrol =pcontrol *(1-pcontrol)
-    print("Expected risk allele frequency:")
-    print(" - In cases: {:.3f}".format(pcase))
-    print(" - In controls: {:.3f}".format(pcontrol))
-
-    num= (pcase - pcontrol)
-    den= np.sqrt( (vcase/scase +  vcontrol/scontrol)*0.5 )
-    u = num / den
-
-    c = ss.norm.isf(sig_level/2)
-    power = 1 - ss.norm.cdf(c-u) + ss.norm.cdf(-c-u)
-    print("Expected power: {:.3f}".format(power))
-    return power
+    if mode=="b":
+        print("Input settings:{}".format(daf))
+        print(" -Number of cases:{}".format(scase))
+        print(" -Number of controls:{}".format(scontrol))
+        print(" -Risk allele OR:{:.3f}".format(genotype_or))
+        print(" -Disease prevalence:{:.3f}".format(prevalence))
+        print(" -Risk allele frequency: {:.3f}".format(daf))
+        print(" -Significance level: {:.3e}".format(sig_level))
+        # 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)
+        bbf = (1- daf)**2
+
+        # additive
+        x = [ 2*genotype_or-1, genotype_or, 1 ] 
+
+        aap= x[0] * prevalence / (x[0]*aaf + x[1]*abf + x[2]*bbf)
+        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)
+        print("Probability of disease :")
+        print(" - Individuals with AA genotype: {:.3f}".format(aap))
+        print(" - Individuals with AB genotype: {:.3f}".format(abp))
+        print(" - Individuals with BB genotype: {:.3f}".format(bbp))
+        
+        pcase= (aap * aaf + abp * abf*0.5) / prevalence
+        pcontrol=((1-aap )* aaf + (1-abp )* abf*0.5) / (1 - prevalence)
+
+        vcase = pcase *(1-pcase)
+        vcontrol =pcontrol *(1-pcontrol)
+        print("Expected risk allele frequency:")
+        print(" - In cases: {:.3f}".format(pcase))
+        print(" - In controls: {:.3f}".format(pcontrol))
+
+        num= (pcase - pcontrol)
+        den= np.sqrt( (vcase/scase +  vcontrol/scontrol)*0.5 )
+        u = num / den
+
+        c = ss.norm.isf(sig_level/2)
+        power = 1 - ss.norm.cdf(c-u) + ss.norm.cdf(-c-u)
+        print("Expected power: {:.3f}".format(power))
+
+    elif mode=="q":
+        if verbose:
+            log.write("Significance level: {}".format(sig_level))
+            log.write("EAF: {}".format(eaf))
+            log.write("BETA: {}".format(beta))
+            log.write("N: {}".format(n))
+            log.write("H2: {}".format(2*eaf*(1-eaf)*(beta**2)))
+            c = ss.chi2.isf(sig_level/2,df=1)
+            NCP = n * 2*eaf*(1-eaf)*(beta**2)/vary
+            power = 1 - ss.ncx2.cdf(c,df=1,nc=NCP)
+
+    return power
+
+def get_beta(
+              mode="b",
+              t=0,
+              genotype_or=1.3 ,
+              eaf=0.1,
+              n=10000,
+              scase= 2000,
+              scontrol= 15000,
+              prevalence= 0.15,
+              daf = 0.2,
+              sig_level= 5e-8,
+              vary=1,
+              log=Log(),
+              verbose=True,
+              n_matrix=500
+             ):
+    if mode=="q":
+        if t >0:
+            def calculate_power_single(
+                                beta, 
+                                eaf, 
+                                n, 
+                                t, 
+                                sig_level=5e-8,vary=1):
+                
+                c = ss.chi2.isf(sig_level/2,df=1)
+                h2 = 2*eaf*(1-eaf)*(beta**2)
+                NCP = n * h2/vary
+                power = 1 - ss.ncx2.cdf(c,df=1,nc=NCP)
+                return power
+            
+            matrix = np.zeros((n_matrix,n_matrix),dtype=float)
+            eafs = np.linspace(0.5,0.0001,n_matrix)
+            betas =  np.linspace(0.0001,10,n_matrix)
+            
+            for i in range(n_matrix):
+                    matrix[i,] = calculate_power_single(beta=betas,eaf=eafs[i],n=n,t=t)
+            
+
+            i,j=1,1
+            eaf_beta = []
+            while i<n_matrix-1 and j<n_matrix-1:
+                if matrix[i,j] < t:
+                    j+=1
+                else:
+                    i+=1
+                    eaf_beta.append((eafs[i],betas[j]))
+        return pd.DataFrame(eaf_beta)

gwaslab/mqqplot.py

@@ -322,7 +322,7 @@ def mqqplot(insumstats,
     
     # CHR and POS ########################################################################
     # chrom and pos exists && (m || r mode)
-    if (chrom is not None) and (pos is not None) and (("m" in mode) or ("r" in mode)):
+    if (chrom is not None) and (pos is not None) and (("qq" in mode) or ("m" in mode) or ("r" in mode)):
         # when manhattan plot, chrom and pos is needed.
         if chrom in insumstats.columns:
             usecols.append(chrom)

{gwaslab-3.4.15.dist-info → gwaslab-3.4.16.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: gwaslab
-Version: 3.4.15
+Version: 3.4.16
 Summary: A collection of handy tools for GWAS SumStats
 Author-email: Yunye <yunye@gwaslab.com>
 Project-URL: Homepage, https://cloufield.github.io/gwaslab/
@@ -45,7 +45,7 @@ Note: GWASLab is being updated very frequently for now. I will release the first
 ## Install
 
 ```
-pip install gwaslab==3.4.14
+pip install gwaslab==3.4.15
 ```
 
 

{gwaslab-3.4.15.dist-info → gwaslab-3.4.16.dist-info}/RECORD

@@ -3,11 +3,11 @@ gwaslab/Log.py,sha256=ILl6fPs_zZBlS2j1eZtgvWE2HGzON5EYDEGb6GZWzs8,758
 gwaslab/Phenotypes.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gwaslab/Sumstats.py,sha256=f-8yAZiktiQxv-nokoJTUGEBlpxTuWspwu-DZ65XzPc,27929
 gwaslab/Sumstatssummary.py,sha256=Q69702tABxTVk6QIbqQ4k6eSujhXr_MVVQ5saOdEhlk,6367
-gwaslab/__init__.py,sha256=w388QRh8ikkBp3pYwUod85Q6zMC7WrmkwIUTA873uQM,1578
+gwaslab/__init__.py,sha256=NIYPYIYZxqoq06D_yoDnFKjfiguRvBmnk1ojygnaoy4,1669
 gwaslab/annotategenename.py,sha256=SKvvH50jXKGdAy-kPxlFuUkk0q6blDJbeXf4uZjZu1A,1249
 gwaslab/annotateplot.py,sha256=eXo5Qn6rK1f9GX5XqzACPYKoUdoverKRukbbiO60rkE,30261
 gwaslab/calculate_gc.py,sha256=7eJlKzX3TdiBWS8kpAkQe1wbui50ge2cmZgyloyQbJI,2159
-gwaslab/calculate_power.py,sha256=2Id09sB7wtUAezaCF0RToACh2tNvjB4x76w-kuzfudA,1937
+gwaslab/calculate_power.py,sha256=OYychyTqBWR0KbNWyRz6sxNxMaMXETtLrdkXPpuo7zE,4400
 gwaslab/compare_effect.py,sha256=Gl43HqkoXXTDWAT1GNGY0scEc9OdgnsKmzIiWqPAi8I,45945
 gwaslab/config.py,sha256=TP-r-DPhJD3XnRYZbw9bQHXaDIkiRgK8bG9HCt-UaLc,580
 gwaslab/download.py,sha256=vMWB3yY_SSMGjGWLXd0kvKI8lms3si3LJO8zaq7Hw5A,15067
@@ -22,7 +22,7 @@ gwaslab/h2_conversion.py,sha256=79Kc0o1Ik4PsRgS01Loc6wT0uaczpWoZDJ2nSq2_bvI,2377
 gwaslab/meta.py,sha256=kIDfLSC2TFidCX_yJdNk2ITfBhWtfrd2J-Q2a_FODns,2191
 gwaslab/metaanalysis.py,sha256=MuFaxzj9vnebut4pWUg29pGxMcPzWcqbEm3q7t-BUpQ,6770
 gwaslab/miamiplot.py,sha256=A0BFw_hDH9EGasebw9rbgcnugvrrW3TESMYSTVF8H6Q,26245
-gwaslab/mqqplot.py,sha256=WK2e3u_-DRSDkLN2Yp8F_Az5SuRAoxzc6Am-i8feahQ,41328
+gwaslab/mqqplot.py,sha256=KWHQZ6ZSB7ty3wYap0nj47-OdfuS6QkyRyH0wiLO9gw,41346
 gwaslab/plotrg.py,sha256=X7_45qiDdbv06xjdpcS3W6K55unnsD5Yvp-M68OCObg,13544
 gwaslab/preformat_input.py,sha256=5e83HENakrR7bwI6O2iGNh-RBGA_D9V_dkry38PAaKU,17944
 gwaslab/processdata.py,sha256=hJQNkrR4IOLw1pZK_OA1v1iNwOmhvlCW9meGnpP5vfU,573
@@ -35,6 +35,7 @@ gwaslab/rsID2chrpos.py,sha256=-pKhY654zS4uULW7FP4yGHNy3e9Wc2ujc1VLBaoUO5A,6564
 gwaslab/textreposition.py,sha256=yHoWhQPfSKVfEjlqPAs6wexRz3EfGhqfP7e6WeqO0xM,2485
 gwaslab/to_formats.py,sha256=LDpYJjQj2gmvgVYXVcHwvqWgsC68hbWOVHepiEb9YEc,20357
 gwaslab/to_pickle.py,sha256=OOoLXiC0Wemyb7jwyeUNU7xMJTAOUT2XDXrd-sxUFFg,1175
+gwaslab/trumpetplot.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 gwaslab/vchangestatus.py,sha256=id3oUsnMrZth7E-jTQHUfNx2UE0b1ABLO9ajbuTMcjY,1527
 gwaslab/version.py,sha256=JQUDzBC67HUiB7dZaEpDZDRqNGRDkMUgxDUmOSytvcA,386
 gwaslab/winnerscurse.py,sha256=o7CjUwLp4B_60yrntV76ESNaoPcl-HPOzSw9cYWzpW4,2050
@@ -46,8 +47,8 @@ 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-3.4.15.dist-info/LICENSE,sha256=GhLOU_1UDEKeOacYhsRN_m9u-eIuVTazSndZPeNcTZA,1066
-gwaslab-3.4.15.dist-info/METADATA,sha256=H3fPWUSP8Lj4dUlmkMPQjcMW5IY2cvbCCPCkgwZngbo,6350
-gwaslab-3.4.15.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
-gwaslab-3.4.15.dist-info/top_level.txt,sha256=PyY6hWtrALpv2MAN3kjkIAzJNmmBTH5a2risz9KwH08,8
-gwaslab-3.4.15.dist-info/RECORD,,
+gwaslab-3.4.16.dist-info/LICENSE,sha256=GhLOU_1UDEKeOacYhsRN_m9u-eIuVTazSndZPeNcTZA,1066
+gwaslab-3.4.16.dist-info/METADATA,sha256=py8Go066oH_HpLAoipmirmc3w5QNtJwToJT68zvhp-I,6350
+gwaslab-3.4.16.dist-info/WHEEL,sha256=pkctZYzUS4AYVn6dJ-7367OJZivF2e8RA9b_ZBjif18,92
+gwaslab-3.4.16.dist-info/top_level.txt,sha256=PyY6hWtrALpv2MAN3kjkIAzJNmmBTH5a2risz9KwH08,8
+gwaslab-3.4.16.dist-info/RECORD,,