def run(FILE):
FHW = open(FILE, "w")
X = load.grab_specific_region("chr1",6229860,6303055, SHOW=False, bins=1000 )
print min(X[:,0]), max(X[:,0])
FHW.write("#chr1,6229860,6303055\n")
FHW.write("~forward\n")
for i in range(X.shape[0]):
FHW.write(str(X[i,0]) + "," + str(X[i,1]) + "\n")
FHW.write("~reverse\n")
for i in range(X.shape[0]):
FHW.write(str(X[i,0]) + "," + str(X[i,2]) + "\n")
X = simulate.runOne(mu=0, s=0.1, l=3, lr=100, ll=-50, we=0.5,wl=0.25, wr=0.25, pie=0.5, pil=0.1, pir=0.9,
N=1000, SHOW=False, bins=1000, noise=True )
X[:,0]+=6303055
X[:,0]*=100.
st, sp = X[0,0], X[-1,0]
print st, sp
FHW.write("#chrN,"+str(st) + "," + str(sp) + "\n")
FHW.write("~forward\n")
for i in range(X.shape[0]):
FHW.write(str(X[i,0]) + "," + str(X[i,1]) + "\n")
FHW.write("~reverse\n")
for i in range(X.shape[0]):
FHW.write(str(X[i,0]) + "," + str(X[i,2]) + "\n")
FHW.close()
coverage_scores = window(X,std=std,lam=lam,step_size=1)
bayes_ks = bayes_factor(X,std=std,lam=lam,step_size=1)
hybrid = center(coverage_scores, bayes_ks)
starts = find_peaks([(x,y) for x,y in zip(np.linspace(X[0,0], X[-1,0], len(hybrid)), hybrid)])
return coverage_scores, bayes_ks, hybrid, starts
def sample(X, k, std=1, lam=0.1):
coverage_scores, bayes_ks, hybrid, starts = compute_possible_EM_starts(X,std=std, lam=lam)
keeps = list()
for i in range(k):
j = np.random.geometric(0.8)-1
keeps.append(starts[j][0])
starts = starts[:j] + starts[j+1:]
return keeps
if __name__=="__main__":
X = load.grab_specific_region("chr1",6229860,6303055, SHOW=True, bins=300 )
X[:,0]/=100.
X[:,0]-=X[0,0]
coverage_scores, bayes_ks, hybrid, starts = compute_possible_EM_starts(X,std=1,lam=0.1)
draw(X, coverage_scores, bayes_ks, hybrid,starts)
clf = model.EMGU(noise=True, K=3,noise_max=0.01,
moveUniformSupport=5,
max_it=50,seed=True)
clf.fit(X)
clf.draw(X)
ax.set_xlabel("Relative Genomic Position")
plt.savefig("/Users/joazofeifa/Lab/Article_drafts/EMG_paper/images/example_gene_fig.svg")
plt.show()
def write_out(X, OUT=""):
FHW = open(OUT, "w")
for i in range(X.shape[0]):
FHW.write(str(X[i,0])+","+str(X[i,1])+","+str(X[i,2])+"\n")
FHW.close()
def load_IN(FILE):
L = list()
with open(FILE) as FH:
for line in FH:
x,y,r = line.strip("\n").split(",")
L.append((float(x), float(y), float(r)))
return np.array(L)
if __name__=="__main__":
WRITE = False
OUT = "/Users/joazofeifa/Lab/Article_drafts/EMG_paper/files/Example_Gene.csv"
if WRITE:
X = load.grab_specific_region("chr1",8012007, 8033978,
pos_file="/Users/joazofeifa//Lab/gro_seq_files/HCT116/bed_graph_files/DMSO2_3.pos.BedGraph",
neg_file="/Users/joazofeifa//Lab/gro_seq_files/HCT116/bed_graph_files/DMSO2_3.neg.BedGraph",
SHOW =False, bins=300)
X[:,0]-=X[0,0]
X[:,0]/=100.
write_out(X, OUT=OUT)
X = load_IN(OUT)
draw(X)
if __name__ == "__main__":
IN = "/Users/joazofeifa/Lab/gro_seq_files/HCT116/bed_graph_files/"
#chr1:87,691,254-87,695,004
#88,319,575-88320266
#92,308,146-92,315,100
#62,182,362-62,198,443
#8,246,915-8,255,824
#chr1:3,233,790-3,239,961
#chr1:1,163,801-1,175,755
#chr1:1,240,585-1,248,496
#1,243,262-1,251,173
#chr1:1,090,956-1,114,133
#chr1:1,087,608-1,108,057
X = load.grab_specific_region("chr1",1087608,1108057, SHOW=False, bins=100,
pos_file=IN+"DMSO2_3.pos.BedGraph", neg_file=IN+"DMSO2_3.neg.BedGraph" )
X[:,0]-=min(X[:,0])
scale = 100
window = 1000
X[:,0]/=scale
run_MM(X, window=window, scale=scale )
N = sum(Y)
XS = sum([ X[i]*Y[i] for i in range(len(X))])
mean = XS/N
var = sum([pow(X[i] - mean,2)*Y[i] for i in range(len(X))]) / N
X2 = sum([pow(X[i],2)*Y[i] for i in range(len(X))])
print var, (X2 - 2*mean*XS + pow(mean,2)*N) /N
if __name__ == "__main__":
IN = "/Users/joazofeifa/Lab/gro_seq_files/HCT116/bed_graph_files/"
#chr1:87,691,254-87,695,004
#88,319,575-88320266
#92,308,146-92,315,100
#62,182,362-62,198,443
#8,246,915-8,255,824
#chr1:3,233,790-3,239,961
#chr1:1,013,872-1,017,272
#934,235-937,997
#1,206,352-1,213,240
#836,632-843,542
#1,091,333-1,096,157
#chr1:162,105,107-162,113,041
#25,681-33,615
X = load.grab_specific_region("chr1",162105107,162113041, SHOW=False, bins=500,
pos_file=IN+"DMSO2_3.pos.BedGraph", neg_file=IN+"DMSO2_3.neg.BedGraph" )
X[:,0]-=min(X[:,0])
scale = 100
window = 500
X[:,0]/=scale
run_MM(X, window=window, scale=scale )
if __name__ == "__main__":
#==================================
#testing MAP-EM procedure
# X = simulate.runOne(mu=0, s=1, l=10, lr=100, ll=-50, we=0.5,wl=0.25, wr=0.25, pie=0.5, pil=0.1, pir=0.9,
# N=1000, SHOW=False, bins=300, noise=False, foot_print=10 )
# chr1:20,984,647-20,991,448
#chr1:836,835-843,549
#chr1:539,399-542,484
#chr3:15,684,556-15,692,636
#chr2:10,420,826-10,462,048
WRITE = False
if WRITE:
X = load.grab_specific_region("chr2",10420826, 10433237,
pos_file="/Users/joazofeifa//Lab/gro_seq_files/HCT116/bed_graph_files/DMSO2_3.pos.BedGraph",
neg_file="/Users/joazofeifa//Lab/gro_seq_files/HCT116/bed_graph_files/DMSO2_3.neg.BedGraph",
SHOW =False, bins=1000)
X[:,0]-=X[0,0]
X[:,0]/=100.
FHW = open("/Users/joazofeifa/test.bed", "w")
for i in range(X.shape[0]):
FHW.write(str(X[i,0]) + "\t" + str(X[i,1]) + "\t" + str(X[i,2]) + "\n")
FHW.close()
X = list()
with open("/Users/joazofeifa/test.bed") as FH:
for line in FH:
x,y,z = [float(x) for x in line.strip("\n").split("\t")]
X.append([x,y,z])
X = np.array(X)
def sample(X, k, std=1, lam=0.1):
coverage_scores, bayes_ks, hybrid, starts = compute_possible_EM_starts(
X, std=std, lam=lam)
keeps = list()
for i in range(k):
j = np.random.geometric(0.8) - 1
keeps.append(starts[j][0])
starts = starts[:j] + starts[j + 1:]
return keeps
if __name__ == "__main__":
X = load.grab_specific_region("chr1",
6229860,
6303055,
SHOW=True,
bins=300)
X[:, 0] /= 100.
X[:, 0] -= X[0, 0]
coverage_scores, bayes_ks, hybrid, starts = compute_possible_EM_starts(
X, std=1, lam=0.1)
draw(X, coverage_scores, bayes_ks, hybrid, starts)
clf = model.EMGU(noise=True,
K=3,
noise_max=0.01,
moveUniformSupport=5,
max_it=50,
seed=True)
clf.fit(X)
#chr1:87,691,254-87,695,004
#88,319,575-88320266
#92,308,146-92,315,100
#62,182,362-62,198,443
#8,246,915-8,255,824
#chr1:3,233,790-3,239,961
#chr1:1,013,872-1,017,272
#934,235-937,997
#1,206,352-1,213,240
#836,632-843,542
#1,091,333-1,096,157
#chr1:162,105,107-162,113,041
#25,681-33,615
#chr1:760,940-764,973
#899,808-905,675
#chr1:4,763,739-4,766,290
#chr1:1,140,801-1,143,549
#:1,200,396-1,202,629
#:936,603-947,066
#1,246,727-1,252,981
#1,137,847-1,145,798
print (1252981 - 1246727)/500.0
X = load.grab_specific_region("chr1",1246727,1252981, SHOW=False, bins=500,
pos_file=IN+"DMSO2_3.pos.BedGraph", neg_file=IN+"DMSO2_3.neg.BedGraph" )
X[:,0]-=min(X[:,0])
scale = 100
window = 500
X[:,0]/=scale
run_MM(X, window=window, scale=scale )