def plotSurf():
from scipy import interpolate
a=pd.read_pickle(utl.outpath+'real/real.maxLikelihoods.df')
idx=(a.s.abs()*a.h.abs()*(a.alt-a.null)).sort_values().index[-1]
R=pd.DataFrame(pd.read_pickle(utl.outpath+'real/real.replicates.df').loc[idx]).T
SH=dta.getSH()
ARGS=[(R,)+sh for sh in SH]
likelihoods=pd.concat(map(mkv.computeLikelihoodReal,ARGS),axis=1);likelihoods.columns.names=['s','h']
fig = plt.figure()
ax = fig.gca(projection='3d')
df=pd.concat([pd.Series(z[1].loc[z[0]].values,index=z[1].loc[z[0]].index,name=z[0]) for z in b.groupby(level=0)],axis=1)
Z=df.values
# Z[Z==Z.min()]=-1e3
X=np.tile(df.index.values[:,None],Z.shape[1])
Y=np.tile(df.columns.values[:,None],Z.shape[0]).T
Z.min()
Z.max()
nn = 401;
xi = np.linspace(-1.0, 2.0, 10);
yi = np.linspace(-0.5, 0.5, nn);
f = interpolate.interp2d(X,Y,Z,kind='cubic')
zi = f(xi, yi)
[xi, yi] = np.meshgrid(xi, yi);
# surf = ax.plot_surface(X, Y, Z, cmap=mpl.cm.autumn)
surf = ax.plot_surface(xi, yi, zi, cmap=mpl.cm.autumn)
fig.colorbar(surf, shrink=0.5, aspect=5)
# surf(xi, yi, zi, 'LineStyle', 'none', 'FaceColor', 'interp')
plt.show()
def computeBaseSFS(recompute=False):
path = utl.outpath + 'real/SFS.F0.df'
if not os.path.exists(path) or recompute:
x0 = dta.getBaseFreq()
import popgen.Estimate as est
sfs = utl.scanGenome(x0, lambda x: est.Estimate.getEstimate(x=x, n=1000, method='all',
selectionPredictor=True)).apply(
lambda x: pd.Series(x[0]), axis=1)
sfs.to_pickle(path)
return sfs
else:
return pd.read_pickle(path)
def SNPscan(R,regAlpha,numProcess):
reload(dta)
SH=dta.getSH(sparse=True)
ARGS=[(R,)+sh for sh in SH]
print pd.DataFrame(SH)
print R
if numProcess==1:
likelihoods=pd.concat(map(mkv.computeLikelihoodReal,ARGS),axis=1);likelihoods.columns.names=['s','h']
else:
pool=Pool(numProcess)
likelihoods=pd.concat(pool.map(mkv.computeLikelihoodReal,ARGS),axis=1);likelihoods.columns.names=['s','h']
pool.terminate()
del ARGS
gc.collect()
likelihoods=mkv.maxLikelihood(likelihoods,regAlpha=regAlpha)
gc.collect()
return likelihoods
def likelihoodWithDifferentN(N=1000,s=0):
T=mkv.computePowerForSandSaveRealData((s,0.5),N=N,save=False)
CD,E=dta.precomputeCDandEmissionsFor(pd.DataFrame(cdi).T,N=N)
return computeLikelihoodReal((CD,E,T)).rename(N)
scan=pd.concat([utl.scanGenome(utl.zpvalgenome(s)).rename('win'),utl.scanGenomeSNP(utl.zpvalgenome(s)).rename('snp')],1)
pplt.Manhattan(scan)
pplt.GenomeChromosomewise(utl.zpvalgenome(utl.scanGenome(utl.zpvalgenome(s))))
pplt.GenomeChromosomewise(utl.zpvalgenome(utl.scanGenome(scores.abs())))
reload(utl)
pplt.GenomeChromosomewise(utl.scanGenomeSNP(utl.zpvalgenome2tail(s)))
scores.sort_values()
pplt.GenomeChromosomewise(utl.scanGenomeSNP(scores.abs(),lambda x: x[x>=x.quantile(0.5)].sum()))
df=pd.concat([scores,s],1);df=pd.concat([df,df.rank()],1,keys=['val','rank']).sort_values(('val','s'))
dfy=pd.concat([df,y],1).dropna()
dfy.sort_values(0)
i=df.index[-1];
cdi=cdAll.loc[i];print cdi.unstack('REP');pplt.plotSiteReal(cdi)
cdiun=cdi.unstack('REP')
CD,E=dta.precomputeCDandEmissionsFor(pd.DataFrame(cdi).T)
h=0.5
reload(mkv)
mkv.computeLikelihoodReal((CD, E, 0, 0.5))
likes=pd.concat(map(lambda x:mkv.computeLikelihoodReal((CD, E, x, 0.5)),S),keys=S).reset_index().iloc[:,[0,-1]].set_index('level_0')[0]
likes[0]
reload(pplt)
plt.figure(figsize=(6,3),dpi=150);plt.subplot(1,2,1);pd.DataFrame(likes).plot(ax=plt.gca());plt.subplot(1,2,2);pplt.plotSiteReal(cdi,ax=plt.gca());print cdi.unstack('REP')
res=res.reset_index().iloc[:,[0,3]];res=res.set_index(res.columns[0]).iloc[:,0]
NN=np.arange(100,1500,100)
def likelihoodWithDifferentN(N=1000,s=0):
T=mkv.computePowerForSandSaveRealData((s,0.5),N=N,save=False)