def main(data_file, output_prefix, degree_file, width, height):
sequences = seq_IO.read_sequences(data_file, additional_params=True, header=True, list_vals=True)
seq_degree = seq_IO.read_sequences(degree_file, additional_params=True, header=True)
degree_frac = defaultdict(list)
for seq, seq_dict in sequences.items():
degree_frac[seq_degree[seq]['Degree']].append(np.mean(seq_dict["Frac"]))
data = [ np.mean(seq_dict["Frac"]) for seq, seq_dict in sequences.items() ]
degree_frac_avg = [ np.mean(list_fracs) for degree, list_fracs in degree_frac.items() ]
degree_frac_std = [ np.std(list_fracs) for degree, list_fracs in degree_frac.items() ]
fig, axarr = pconv.create_ax(1, 1, shx=False, shy=False)
hist.draw_actual_plot(axarr[0,0], data, "", "", normed=False, nbins=30, edgecolor=None, log=False)
#axarr[0,0].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
pconv.save_fig(fig, output_prefix, "hist", width, height, tight=True, size=10)
fig2, axarr2 = pconv.create_ax(1, 1, shx=True, shy=True)
bar.draw_actual_plot(axarr2[0,0], degree_frac_avg, 'g', "", "Degree", "Fraction Shortest Path Uncleaved", tick_label=degree_frac.keys(), yerr=degree_frac_std)
#axarr[0,0].set_ylim([0,1.3])
pconv.save_fig(fig2, output_prefix, "bar", width, height, tight=True, size=10)
def main(epistasis_file):
dict_epistasis = {} #list of list of sequences, where each item represents a label
with open(epistasis_file) as e:
lines = e.readlines()
for l in lines[1:]: #ignore header line
tokens = l.split(',')
#value consists of Starting Ratio, Ending Ratio, Epistasis, Ending Fitness, # of Mutations, list of InterSeqs, list of InterFits, list of InterRatios
if dict_epistasis.get((tokens[2], tokens[0])) is None:
dict_epistasis[(tokens[0],tokens[2])] = [ float(tokens[1]), float(tokens[3]), float(tokens[5]), tokens[4], len(tokens[6::3]), tokens[6::3], tokens[7::3], tokens[8::3] ]
neg_epistasis = [0] * 4
no_epistasis = [0] * 4
pos_epistasis = [0] * 4
n_functional = [0] * 4
n_should_be_functional = [0] * 4
n_total = [0] * 4
for i in xrange(2,6):
ind = i-2
neg_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[2] < -0.000005 and value[4] == i ])
no_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if abs(value[2]) < 0.000005 and value[4] == i ])
pos_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[2] > 0.000005 and value[4] == i ])
n_functional[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[3] == "CLEAVED" and value[4] == i ])
n_should_be_functional[ind] = sum([ 1 for key, value in dict_epistasis.items() if all(v == "CLEAVED" for v in value[6]) and value[4] == i ])
n_total[ind] = float(sum([ 1 for key, value in dict_epistasis.items() if value[4] == i]))
seq_func = set([ key[1] for key,val in dict_epistasis.items() if val[3] == "CLEAVED" ])
seq_pred_func = set([ key[1] for key,val in dict_epistasis.items() if all(v == "CLEAVED" for v in val[6]) ])
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
fig2, axarr2 = pconv.create_ax(1, 1)
artists = []
artists.extend(plot_epi(no_epistasis, n_total, axarr[0,0], "No", color="gray"))
artists.extend(plot_epi(neg_epistasis, n_total, axarr[0,0], "Neg.", bottom=no_epistasis, color="white"))
artists.extend(plot_epi(pos_epistasis, n_total, axarr[0,0], "Pos.", bottom=[no + neg for no, neg in zip(no_epistasis, neg_epistasis)], color="black"))
n_func_frac = [ func/total for func, total in zip(n_functional, n_total) ]
n_pred_frac = [ pred/total for pred, total in zip(n_should_be_functional, n_total) ]
scatterplot.plot_series(axarr2[0,0], [(range(2,6),n_func_frac,"% Cleaved"),(range(2,6),n_pred_frac,"% Pred Cleaved")], "", "Number of Mutations", "Fraction of Total Cases", size=40, connect_dots=True, alpha=1.0)
axarr2[0,0].set_ylim([0,4.0])
fig_venn, axarr_venn = pconv.create_ax(1, 1)
venn2([seq_func, seq_pred_func], set_labels = ["Cleaved", "Pred Cleaved"], ax=axarr_venn[0,0])
lgd = axarr[0,0].legend(artists,["No","Neg.","Pos."], loc="center left", bbox_to_anchor=(1.05, 0.5), borderaxespad=0., prop={'size':9}, ncol=1, fancybox=True)
pconv.save_fig(fig, epistasis_file, "plot", 3, 2.5, tight=False, size=9, extra_artists=lgd)
pconv.save_fig(fig2, epistasis_file, "pred_v_cl", 5, 5, tight=True, size=10)
pconv.save_fig(fig_venn, epistasis_file, "venn", 5, 5, tight=True, size=14)
def plot_dict(dict_to_plot,
samplesize,
output_pre,
suffix,
x_axis,
norm="all"):
fig, axarr = conv.create_ax(len(dict_to_plot), 1, shx=True, shy=True)
for ind, (key, (val, sample_name)) in enumerate(dict_to_plot.items()):
samplesize_list = [
l for k, l in samplesize.items() if k == sample_name
][0] #assume only one item meets that criteria
title = ''.join(key)[0:28]
plot_curve(axarr[0, ind],
val,
samplesize_list,
"",
title,
x_axis,
"Sample Size",
norm=norm)
conv.save_fig(fig,
output_pre + (norm if norm is not None else "none"),
suffix,
len(dict_to_plot) * 4,
4,
tight=True,
size=10)
def main(json_file, output_prefix, metric):
with open(json_file) as data_file:
data = json.load(data_file)
G = json_graph.node_link_graph(data)
metrics = {}
#metrics["degree"] = degree(G)
metrics["closeness"] = closeness_centrality(G).values()
#TODO: add any other metrics here using a similar format to above line.
sequences = {}
cleaved_seq = { key : val for key, val in sequences.items() if val["type"] == "CLEAVED" }
if metric != "metrics":
labels_to_plot = [metric]
else:
labels_to_plot = metrics.keys()
n_to_plot = len(labels_to_plot)
fig, axarr = pconv.create_ax(n_to_plot, 1, shx=False, shy=False)
nbins = 20
for ind, key in enumerate(labels_to_plot):
normed = True
hist.draw_actual_plot(axarr[0,ind], metrics["key"], "", key.capitalize(), normed=normed, nbins=nbins)
axarr[0,ind].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
#pconv.add_legend(axarr[0,ind], location="middle right")
pconv.save_fig(fig, output_prefix, "metrics", n_to_plot*5, 5, tight=True, size=12)
def main(sequence_ratio_file, width, height, pattern, legend):
sequences = seq_IO.read_sequences(sequence_ratio_file, additional_params=True)
shell_data = []
for shell in xrange(1,len(sequences[0])):
shell_data.append([ seq[shell] for seq in sequences ])
avg = []
std = []
label = xrange(1,4)
for sd in shell_data:
avg.append( np.median(sd))
std.append( np.std(sd))
#check if std has to be fixed
#if sum([ 1 for a, s in zip(avg_ratio, std) if a - s < 0 ]):
# min_err = [ a - s if a - s >= 0.0 else 0 for a,s in zip(avg_ratio, std) ]
# max_err = [ a + s for a,s in zip(avg_ratio, std) ]
# err = [min_err, max_err]
#else:
# err = std
err = std
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
bar.draw_actual_plot(axarr[0,0], avg, ['lightsteelblue','lightblue','darkgray'], "", "Shell", "Fraction Cleaved", tick_label=label, yerr = err)
#axarr[0,0].set_ylim([0,1.3])
pconv.save_fig(fig, sequence_ratio_file, "plot", width, height, tight=True, size=10)
def main(list_sequence_names, output_prefix):
list_sequences = [] #list of list of sequences, where each item represents a label
extended_list_sequences = [] #flat list of sequences
labels = [] #labels for list_sequences
for [filename, label] in list_sequence_names:
sequences = seq_IO.read_sequences(filename)
list_sequences.append(sequences)
extended_list_sequences.extend(sequences[:])
labels.append(label)
cleaved_ind = labels.index("CLEAVED")
middle_ind = labels.index("MIDDLE")
uncleaved_ind = labels.index("UNCLEAVED")
frac_uncleaved = {}
frac_cleaved = {}
frac_middle = {}
for seq in list_sequences[cleaved_ind]:
cleaved_seqs = sum([1 for s in list_sequences[cleaved_ind] if conv.hamdist(seq,s) == 1])
uncleaved_seqs = sum([1 for s in list_sequences[uncleaved_ind] if conv.hamdist(seq,s) == 1])
middle_seqs = sum([1 for s in list_sequences[middle_ind] if conv.hamdist(seq,s) == 1])
if cleaved_seqs > 0 or uncleaved_seqs > 0:
total = uncleaved_seqs+middle_seqs+cleaved_seqs
frac_uncleaved[seq] = float(uncleaved_seqs)/total
frac_cleaved[seq] = float(cleaved_seqs)/total
frac_middle[seq] = float(middle_seqs)/total
fig, ax = pconv.create_ax(3, 1)
hist.draw_actual_plot(ax[0,0], frac_cleaved.values(), "Landscape Near Cleaved Sequences", "Fraction of Neighbors Cleaved", log=False, normed=False, nbins=20)
hist.draw_actual_plot(ax[0,1], frac_middle.values(), "Landscape Near Cleaved Sequences", "Fraction of Neighbors Middle", log=False, normed=False, nbins=20)
hist.draw_actual_plot(ax[0,2], frac_uncleaved.values(), "Landscape Near Cleaved Sequences", "Fraction of Neighbors Uncleaved", log=False, normed=False, nbins=20)
pconv.save_fig(fig, output_prefix, "fraction_neighbors", 15, 5, size=10)
def main(sequences_ratio_file):
sequences_ratio = seq_IO.read_sequences(sequences_ratio_file,
additional_params=True)
seq_ratio_dict = [[l[1], l[2], l[3]] for l in (sequences_ratio)]
seq_cleaved_dict = [l[4] for l in sequences_ratio]
seqs = [l[0] for l in sequences_ratio]
avg_ratio = [sum(v) / 3.0 for v in seq_ratio_dict]
min_ratio = [sum(v) / 3.0 - min(v) for v in seq_ratio_dict]
max_ratio = [max(v) - sum(v) / 3.0 for v in seq_ratio_dict]
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
bar.draw_actual_plot(axarr[0, 0],
avg_ratio,
'c',
"",
"Sequence",
"FLAG/HA Ratio",
tick_label=seqs,
yerr=[min_ratio, max_ratio])
pconv.save_fig(fig,
sequences_ratio_file,
"plot",
4,
4,
tight=True,
size=12)
def plot_dict(dict_to_plot, samplesize, output_pre, suffix, x_axis, norm="all"):
fig, axarr = conv.create_ax(len(dict_to_plot), 1, shx=True, shy=True)
for ind, (key, (val,sample_name)) in enumerate(dict_to_plot.items()):
samplesize_list = [ l for k, l in samplesize.items() if k == sample_name ][0] #assume only one item meets that criteria
title = ''.join(key)[0:28]
plot_curve(axarr[0,ind], val, samplesize_list, "", title, x_axis, "Sample Size", norm=norm)
conv.save_fig(fig, output_pre + (norm if norm is not None else "none"), suffix, len(dict_to_plot)*4, 4, tight=True, size=10)
def main(list_sequence_names, canonical_list, output_prefix, func_labels,
unfunc_labels):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_list)
for canonical in canonical_list_seq:
dict_sequences = {}
for [filename, label] in list_sequence_names:
sequences = seq_IO.read_sequences(filename)
distances = [conv.hamdist(seq, canonical) for seq in sequences]
dict_sequences[label] = {
i: sum([d for d in distances if d == i])
for i in xrange(1, 6)
}
x = []
y = []
for i in xrange(1, 6):
func = 0.0
unfunc = 0.0
for label, dict_sums in dict_sequences.items():
if label in func_labels:
func = func + dict_sums[i]
elif label in unfunc_labels:
unfunc = unfunc + dict_sums[i]
if unfunc != 0:
x.append(i)
y.append(func / (func + unfunc))
print x
print y
series.append([x, y, canonical])
fig, ax = pconv.create_ax(1, 1)
scatterplot.plot_series(ax[0, 0],
series,
title="",
x_axis="# of Mutations",
y_axis="Fraction of Variants that are Functional",
alpha=1.0,
connect_dots=True,
size=30,
edgecolors='k')
ax[0, 0].set_xlim(xmin=1, xmax=5)
ax[0, 0].set_xticks(xrange(1, 6))
pconv.save_fig(fig,
output_prefix,
canonical + "_fraction_func_mutant",
6,
6,
size=15)
def main(data_file, title, output_prefix):
sequences = seq_IO.read_sequences(data_file, additional_params=True, header=True)
data = [ seq_dict["Degree"] for seq, seq_dict in sequences.items() ]
fig, axarr = pconv.create_ax(1, 1, shx=False, shy=False)
hist.draw_actual_plot(axarr[0,0], data, "", title.capitalize(), normed=True, nbins=30, edgecolor=None, log=False)
#axarr[0,0].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
pconv.save_fig(fig, output_prefix, title, 5, 5, tight=True, size=10)
def main(list_sequence_names, output_prefix):
sequence_list = []
labels = []
for [filename, label] in list_sequence_names:
sequence_list.append(set(seq_IO.read_sequences(filename)))
labels.append(label)
fig, ax = pconv.create_ax(1, 1)
venn3(sequence_list, set_labels = labels, ax=ax[0,0])
pconv.save_fig(fig, output_prefix, '_'.join(labels)+"_venn", 10, 10, size=12)
def main(sequences_ratio_file):
sequences_ratio = seq_IO.read_sequences(sequences_ratio_file, additional_params=True)
seq_ratio_dict = [ [l[1],l[2],l[3]] for l in (sequences_ratio) ]
seq_cleaved_dict = [ l[4] for l in sequences_ratio ]
seqs = [ l[0] for l in sequences_ratio ]
avg_ratio = [ sum(v)/3.0 for v in seq_ratio_dict]
min_ratio = [ sum(v)/3.0 - min(v) for v in seq_ratio_dict]
max_ratio = [ max(v)-sum(v)/3.0 for v in seq_ratio_dict]
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
bar.draw_actual_plot(axarr[0,0], avg_ratio, 'c', "", "Sequence", "FLAG/HA Ratio", tick_label=seqs, yerr = [min_ratio, max_ratio] )
pconv.save_fig(fig, sequences_ratio_file, "plot", 4, 4, tight=True, size=12)
def main(list_sequence_names, output_prefix):
lines = []
temp_dict = { "CLEAVED" : {}, "UNCLEAVED" : {}, "MIDDLE" : {} }
for [filename, label, sample] in list_sequence_names:
sequences = seq_IO.read_sequences(filename)
temp_dict[label][sample] = len(sequences)
lines.append(([ val for k, val in sorted(temp_dict["CLEAVED"].items()) ], "CLEAVED") )
lines.append(([ val for k, val in sorted(temp_dict["MIDDLE"].items()) ], "MIDDLE") )
lines.append(([ val for k, val in sorted(temp_dict["UNCLEAVED"].items()) ], "UNCLEAVED") )
fig, ax = pconv.create_ax(1, 1)
bar.plot_series( ax[0,0], lines, title="", x_axis="Variant Name", y_axis="Number of Substrate Sequences Sampled", tick_label=sorted(temp_dict["CLEAVED"].keys()))
pconv.save_fig(fig, output_prefix, "cleaved_uncleaved_middle", 6, 6, tight=True, size=10)
def main(list_nodes, output_prefix, metric, create_keys=False):
if not create_keys:
sequences = seq_IO.read_sequences(list_nodes, additional_params=True, header=True)
else:
sequences = seq_IO.read_sequences(list_nodes, additional_params=True, header=True, create_keys=True)
cleaved_seq = { key : val for key, val in sequences.items() if val["type"] == "CLEAVED" }
middle_seq = { key : val for key, val in sequences.items() if val["type"] == "MIDDLE" }
uncleaved_seq = { key : val for key, val in sequences.items() if val["type"] == "UNCLEAVED" }
print len(cleaved_seq)
if metric == "metrics":
labels_non_plot = ["label", "fitness", "type", "canonical", "timeset"]
#labels_to_plot = sorted([ key for key in sequences["YNYIN"].keys() if key not in labels_non_plot ] + ["Fraction_Cleaved"])
labels_to_plot = sorted([ key for key in sequences["YNYIN"].keys() if key not in labels_non_plot ])
else:
labels_to_plot = [metric]
n_to_plot = len(labels_to_plot)
fig, axarr = pconv.create_ax(n_to_plot, 1, shx=False, shy=False)
nbins = 10
for ind, key in enumerate(labels_to_plot):
if key == "pageranks":
log = True
else:
log = False
if key == "Fraction_Cleaved":
# data = [ conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), cleaved_seq.keys()).values(),
# conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), middle_seq.keys()).values(),
# conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), uncleaved_seq.keys()).values()]
normed = True
else:
data = [ get_data_from_dict(cleaved_seq, key), get_data_from_dict(middle_seq, key), get_data_from_dict(uncleaved_seq, key) ]
normed = True
print key
hist.draw_actual_plot(axarr[0,ind], data, "", key.capitalize(), log=log, normed=normed, label=["Cleaved", "Middle", "Uncleaved"], nbins=nbins)
axarr[0,ind].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
#pconv.add_legend(axarr[0,ind], location="middle right")
pconv.save_fig(fig, output_prefix, metric, n_to_plot*2.5, 2.5, tight=True, size=9)
def gen_plots(c, r, ax2, output_pre, dirname, st=""):
fig, axarr = conv.create_ax(1, 31, shx=True, shy=True)
counters = []
coeffs = []
pvals = []
for i in xrange(1,31):
m = combine_counts_ratios(c, r, i, i+9, st=st)
coeff, pval = find_coeff_pval(m, axarr[i-1,0], "Sliding Window: {0} to {1}".format(i, i+9) )
print i, coeff, pval
counters.append(i)
coeffs.append(coeff)
pvals.append(pval)
plot_coeff_pval(axarr[30,0], counters, coeffs, pvals)
suffix = os.path.normpath(dirname).split(os.sep)[-3]
conv.save_fig(fig, output_pre + "correlation_plot.txt", "{0}_{1}".format(suffix,st), 4, 20*4)
plot_coeff_pval(ax2, counters, coeffs, pvals, suffix)
def main(ratios_file1, ratios_file2, output_pre, use_sel):
if use_sel:
counts1_dict = read_ratios(ratios_file1)[2]
counts2_dict = read_ratios(ratios_file2)[2]
title1 = os.path.basename(ratios_file1).split("_")[1]
title2 = os.path.basename(ratios_file2).split("_")[1]
else:
counts1_dict = read_ratios(ratios_file1)[1]
counts2_dict = read_ratios(ratios_file2)[1]
title1 = os.path.basename(ratios_file1).split("_")[6]
title2 = os.path.basename(ratios_file2).split("_")[6]
fig, axarr = conv.create_ax(2, 1)
c1, c2 = common_points(counts1_dict, counts2_dict)
plot_corr(c1, c2, axarr[0,0], title1, title2)
c1, c2 = common_points(counts1_dict, counts2_dict, filtered=True)
plot_corr(c1, c2, axarr[0,1], title1, title2)
conv.save_fig(fig, output_pre + "/corrcounts.txt", title1 + "_" + title2, 20, 10, tight=True)
def main(args):
#read in and rename arguments
title1 = os.path.basename(args.input_dir_1)
title2 = os.path.basename(args.input_dir_2)
d1, n1 = scorefileparse.read_dec_nat(args.input_dir_1, [], args.scoretype1, True)
d2, n2 = scorefileparse.read_dec_nat(args.input_dir_2, [], args.scoretype2, True)
dec1 = scorefileparse.filter_pdbs_by_rmsd(d1, args.rmsd_cutoff)
nat1 = scorefileparse.filter_pdbs_by_rmsd(n1, args.rmsd_cutoff)
dec2 = scorefileparse.filter_pdbs_by_rmsd(d2, args.rmsd_cutoff)
nat2 = scorefileparse.filter_pdbs_by_rmsd(n2, args.rmsd_cutoff)
dec_norm1 = scorefileparse.norm_pdbs(dec1)
nat_norm1 = scorefileparse.norm_pdbs(nat1,dec1)
dec_norm2 = scorefileparse.norm_pdbs(dec2)
nat_norm2 = scorefileparse.norm_pdbs(nat2,dec2)
[dec_inter1, nat_inter1, dec_inter2, nat_inter2] = scorefileparse.pdbs_intersect([dec_norm1, nat_norm1, dec_norm2, nat_norm2])
[dec_inter1, dec_inter2] = scorefileparse.pdbs_scores_intersect([dec_inter1, dec_inter2])
[nat_inter1, nat_inter2] = scorefileparse.pdbs_scores_intersect([nat_inter1, nat_inter2])
dec_filt1 = scorefileparse.filter_norm_pdbs(dec_norm1)
nat_filt1 = scorefileparse.filter_norm_pdbs(nat_norm1)
dec_filt2 = scorefileparse.filter_norm_pdbs(dec_norm2)
nat_filt2 = scorefileparse.filter_norm_pdbs(nat_norm2)
[dec_finter1, dec_finter2] = scorefileparse.pdbs_scores_intersect([dec_filt1, dec_filt2])
[nat_finter1, nat_finter2] = scorefileparse.pdbs_scores_intersect([nat_filt1, nat_filt2])
fig, axarr = conv.create_ax(2, len(dec_inter1))
for x_ind,pdb in enumerate(sorted(dec_inter1.keys())):
ax = axarr[x_ind, 0]
plot(dec_inter1, dec_inter2, nat_inter1, nat_inter2, ax, pdb, title1, title2)
ax = axarr[x_ind, 1]
plot(dec_finter1, dec_finter2, nat_finter1, nat_finter2, ax, pdb, title1, title2)
def main(args):
#read in and rename arguments
inp_dir=args[1]
scoretype=args[2]
dec, nat = scorefileparse.read_dec_nat(inp_dir, [], scoretype)
disc = discparse.read_dir(inp_dir)
dec_norm = scorefileparse.norm_pdbs(dec)
nat_norm = scorefileparse.norm_pdbs(nat,dec)
[dec_inter, nat_inter, disc_inter] = scorefileparse.pdbs_intersect([dec_norm, nat_norm, disc])
#labels = ["Average","1.0","1.5","2.0","2.5","3.0","4.0","6.0"]
labels = ["Average"]
energy_gap = [[] for l in labels]
avg_disc = [[] for l in labels]
for pdb in dec_inter.keys():
for ind in xrange(0,len(labels)):
lowest_dec = min([ e[0] for e in dec_inter[pdb].values() ])
lowest_nat = min([ n[0] for n in nat_inter[pdb].values() if n[1] < 2.0 ])
energy_gap[ind].append(lowest_nat - lowest_dec)
avg_disc[ind].append(disc_inter[pdb][0])
fig, axarr = conv.create_ax(len(labels), 1)
for x_ind,l in enumerate(labels):
ax = axarr[0,x_ind]
scatterplot.draw_actual_plot(ax, avg_disc[x_ind], energy_gap[x_ind], [], l,"Disc","Energy Gap")
scatterplot.plot_regression(ax, avg_disc[x_ind], energy_gap[x_ind], False, False)
title = os.path.basename(inp_dir)
filename=inp_dir + "/test.txt"
conv.save_fig(fig, filename, "disc_v_egap", len(labels)*3, 4)
def gen_plots(c, r, ax2, output_pre, dirname, st=""):
fig, axarr = conv.create_ax(1, 31, shx=True, shy=True)
counters = []
coeffs = []
pvals = []
for i in xrange(1, 31):
m = combine_counts_ratios(c, r, i, i + 9, st=st)
coeff, pval = find_coeff_pval(
m, axarr[i - 1, 0], "Sliding Window: {0} to {1}".format(i, i + 9))
print i, coeff, pval
counters.append(i)
coeffs.append(coeff)
pvals.append(pval)
plot_coeff_pval(axarr[30, 0], counters, coeffs, pvals)
suffix = os.path.normpath(dirname).split(os.sep)[-3]
conv.save_fig(fig, output_pre + "correlation_plot.txt",
"{0}_{1}".format(suffix, st), 4, 20 * 4)
plot_coeff_pval(ax2, counters, coeffs, pvals, suffix)
def process_dir(dirnames, unsel, output_pre):
fig_all, axarr_all = conv.create_ax(len(dirnames), 3, shx=True, shy=True)
for ind,dirname in enumerate(dirnames):
print dirname
counts_fn = dirname + '/counts_' + unsel + '*_PRO_qc'
ratios_fn = dirname + '/ratios_*_PRO_qc'
c_fn = glob.glob(counts_fn)
if len(c_fn) == 0:
c_fn = None
else:
c_fn = c_fn[0]
r_fn = glob.glob(ratios_fn)[0]
c,r = read_files(c_fn, r_fn)
gen_plots(c, r, axarr_all[0,ind], output_pre, dirname, st="")
gen_plots(c, r, axarr_all[1,ind], output_pre, dirname, st="mean")
gen_plots(c, r, axarr_all[2,ind], output_pre, dirname, st="median")
conv.save_fig(fig_all, output_pre + "all_coeff_pval.txt", "", 4*len(dirnames), 12)
def process_dir(dirnames, unsel, output_pre):
fig_all, axarr_all = conv.create_ax(len(dirnames), 3, shx=True, shy=True)
for ind, dirname in enumerate(dirnames):
print dirname
counts_fn = dirname + '/counts_' + unsel + '*_PRO_qc'
ratios_fn = dirname + '/ratios_*_PRO_qc'
c_fn = glob.glob(counts_fn)
if len(c_fn) == 0:
c_fn = None
else:
c_fn = c_fn[0]
r_fn = glob.glob(ratios_fn)[0]
c, r = read_files(c_fn, r_fn)
gen_plots(c, r, axarr_all[0, ind], output_pre, dirname, st="")
gen_plots(c, r, axarr_all[1, ind], output_pre, dirname, st="mean")
gen_plots(c, r, axarr_all[2, ind], output_pre, dirname, st="median")
conv.save_fig(fig_all, output_pre + "all_coeff_pval.txt", "",
4 * len(dirnames), 12)
def main(list_sequence_names, canonical_list, output_prefix ):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_list)
cleaved_seqs = seq_IO.read_sequences( [ s for s,l in list_sequence_names if l == "CLEAVED" ][0] )
uncleaved_seqs = seq_IO.read_sequences( [ s for s,l in list_sequence_names if l == "UNCLEAVED" ][0] )
min_dist = []
avg_dist = []
max_dist = []
for seq in cleaved_seqs:
distances = [ conv.hamdist(seq, unc) for unc in uncleaved_seqs ]
min_dist.append(min(distances))
avg_dist.append(numpy.mean(distances))
max_dist.append(max(distances))
if seq in canonical_list_seq:
print seq
print min_dist[-1]
print avg_dist[-1]
print max_dist[-1]
fig, ax = pconv.create_ax(1, 3)
hist.draw_actual_plot(ax[0,0], min_dist, "Min. Distance from Boundary", "Minimum Distances", log=False, normed=True, label=None, nbins=15, stacked=False)
hist.draw_actual_plot(ax[1,0], avg_dist, "Avg. Distance from Boundary", "Average Distances", log=False, normed=True, label=None, nbins=15, stacked=False)
hist.draw_actual_plot(ax[2,0], max_dist, "Max. Distance from Boundary", "Maximum Distances", log=False, normed=True, label=None, nbins=15, stacked=False)
#ax[0,0].set_xlim(xmin=1,xmax=5)
#ax[0,0].set_xticks(xrange(1,6))
pconv.save_fig(fig, output_prefix, "dist_from_bounds", 18, 6, size=15)
def main(list_sequence_names, canonical_list, output_prefix, func_labels, unfunc_labels):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_list)
for canonical in canonical_list_seq:
dict_sequences = {}
for [filename, label] in list_sequence_names:
sequences = seq_IO.read_sequences(filename)
distances = [ conv.hamdist(seq, canonical) for seq in sequences ]
dict_sequences[label] = { i : sum([d for d in distances if d == i]) for i in xrange(1,6) }
x = []
y = []
for i in xrange(1,6):
func=0.0
unfunc=0.0
for label, dict_sums in dict_sequences.items():
if label in func_labels:
func = func + dict_sums[i]
elif label in unfunc_labels:
unfunc = unfunc + dict_sums[i]
if unfunc != 0:
x.append(i)
y.append( func/(func+unfunc) )
print x
print y
series.append([x, y, canonical])
fig, ax = pconv.create_ax(1, 1)
scatterplot.plot_series( ax[0,0], series, title="", x_axis="# of Mutations", y_axis="Fraction of Variants that are Functional", alpha=1.0, connect_dots=True, size=30, edgecolors='k')
ax[0,0].set_xlim(xmin=1,xmax=5)
ax[0,0].set_xticks(xrange(1,6))
pconv.save_fig(fig, output_prefix, canonical + "_fraction_func_mutant", 6, 6, size=15)
def plot(disc_metrics_1, disc_metrics_2, title1, title2, output_pre, add_slash=True):
pdbs = sorted(disc_metrics_1.keys())
n_metrics = len(disc_metrics_1[pdbs[0]])
fig, axarr = conv.create_ax(n_metrics, 1)
for x_ind,metric_name in enumerate(disc_metrics_1[pdbs[0]].keys()):
x = []
y = []
ax = axarr[0,x_ind]
for pdb in pdbs:
x.append(disc_metrics_1[pdb][metric_name])
y.append(disc_metrics_2[pdb][metric_name])
scatterplot.draw_actual_plot(ax, x, y, 'b', metric_name, title1,title2, size=20, edgecolors='k')
scatterplot.plot_regression(ax,x,y,False)
if add_slash:
filename = output_pre + "/" + title1 + "_" + title2 + ".txt"
else:
filename = output_pre + title1 + "_" + title2 + ".txt"
suffix="disc_v_disc"
conv.save_fig(fig, filename, suffix, n_metrics*3, 3, size=9)
def main(data_file, output_prefix, degree_file, width, height):
sequences = seq_IO.read_sequences(data_file,
additional_params=True,
header=True,
list_vals=True)
seq_degree = seq_IO.read_sequences(degree_file,
additional_params=True,
header=True)
degree_frac = defaultdict(list)
for seq, seq_dict in sequences.items():
degree_frac[seq_degree[seq]['Degree']].append(np.mean(
seq_dict["Frac"]))
data = [np.mean(seq_dict["Frac"]) for seq, seq_dict in sequences.items()]
degree_frac_avg = [
np.mean(list_fracs) for degree, list_fracs in degree_frac.items()
]
degree_frac_std = [
np.std(list_fracs) for degree, list_fracs in degree_frac.items()
]
fig, axarr = pconv.create_ax(1, 1, shx=False, shy=False)
hist.draw_actual_plot(axarr[0, 0],
data,
"",
"",
normed=False,
nbins=30,
edgecolor=None,
log=False)
#axarr[0,0].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
pconv.save_fig(fig,
output_prefix,
"hist",
width,
height,
tight=True,
size=10)
fig2, axarr2 = pconv.create_ax(1, 1, shx=True, shy=True)
bar.draw_actual_plot(axarr2[0, 0],
degree_frac_avg,
'g',
"",
"Degree",
"Fraction Shortest Path Uncleaved",
tick_label=degree_frac.keys(),
yerr=degree_frac_std)
#axarr[0,0].set_ylim([0,1.3])
pconv.save_fig(fig2,
output_prefix,
"bar",
width,
height,
tight=True,
size=10)
def main(epistasis_file):
dict_epistasis = {
} #list of list of sequences, where each item represents a label
with open(epistasis_file) as e:
lines = e.readlines()
for l in lines[1:]: #ignore header line
tokens = l.split(',')
#value consists of Starting Fitness, Ending_Fitness,Epistasis,List_Seqs_Fitnesses_Intermediates
if dict_epistasis.get((tokens[2], tokens[0])) is None:
dict_epistasis[tokens[0]] = [
tokens[1], tokens[2],
float(tokens[3]), tokens[4::2],
[t.strip() for t in tokens[5::2]]
]
'''
n_functional = [0] * 4
n_should_be_functional = [0] * 4
n_total = [0] * 4
for i in xrange(2,6):
ind = i-2
neg_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[2] < -0.000005 and value[4] == i ])
no_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if abs(value[2]) < 0.000005 and value[4] == i ])
pos_epistasis[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[2] > 0.000005 and value[4] == i ])
n_functional[ind] = sum([ 1 for key, value in dict_epistasis.items() if value[3] == "CLEAVED" and value[4] == i ])
n_should_be_functional[ind] = sum([ 1 for key, value in dict_epistasis.items() if all(v == "CLEAVED" for v in value[6]) and value[4] == i ])
n_total[ind] = float(sum([ 1 for key, value in dict_epistasis.items() if value[4] == i]))
'''
seq_func = set(
[key for key, val in dict_epistasis.items() if val[1] == "CLEAVED"])
seq_pred_func = set([
key for key, val in dict_epistasis.items()
if all(v == "CLEAVED" for v in val[4])
])
seq_unfunc = set(
[key for key, val in dict_epistasis.items() if val[1] == "UNCLEAVED"])
seq_pred_unfunc = set([
key for key, val in dict_epistasis.items()
if any(v == "UNCLEAVED" for v in val[4]) or sum(v == "MIDDLE"
for v in val[4]) == 2
])
seq_midfunc = set(
[key for key, val in dict_epistasis.items() if val[1] == "MIDDLE"])
seq_pred_midfunc = set([
key for key, val in dict_epistasis.items()
if any(v == "MIDDLE" for v in val[4])
])
#fig, axarr = pconv.create_ax(3, 1, shx=True, shy=True)
#fig2, axarr2 = pconv.create_ax(1, 1)
#plot_epi(neg_epistasis, n_total, axarr[0,0], "Negative")
#plot_epi(no_epistasis, n_total, axarr[0,1], "No")
#plot_epi(pos_epistasis, n_total, axarr[0,2], "Positive")
#n_func_frac = [ func/total for func, total in zip(n_functional, n_total) ]
#n_pred_frac = [ pred/total for pred, total in zip(n_should_be_functional, n_total) ]
#scatterplot.plot_series(axarr2[0,0], [(range(2,6),n_func_frac,"% Cleaved"),(range(2,6),n_pred_frac,"% Pred Cleaved")], "", "Number of Mutations", "Fraction of Total Cases", size=40, connect_dots=True, alpha=1.0)
#axarr2[0,0].set_ylim([0,1.0])
fig_venn, axarr_venn = pconv.create_ax(1, 1)
fig_vennun, axarr_vennun = pconv.create_ax(1, 1)
fig_vennmid, axarr_vennmid = pconv.create_ax(1, 1)
venn2([seq_func, seq_pred_func],
set_labels=["Cleaved", "Pred Cleaved"],
ax=axarr_venn[0, 0])
venn2([seq_unfunc, seq_pred_unfunc],
set_labels=["Uncleaved", "Pred Uncleaved"],
ax=axarr_vennun[0, 0])
venn2([seq_midfunc, seq_pred_midfunc],
set_labels=["Middle", "Pred Middle"],
ax=axarr_vennmid[0, 0])
#pconv.save_fig(fig, epistasis_file, "plot", 12, 4, tight=True, size=12)
#pconv.save_fig(fig2, epistasis_file, "pred_v_cl", 5, 5, tight=True, size=10)
pconv.save_fig(fig_venn,
epistasis_file,
"venn",
5,
5,
tight=False,
size=14)
pconv.save_fig(fig_vennun,
epistasis_file,
"vennun",
5,
5,
tight=False,
size=14)
pconv.save_fig(fig_vennmid,
epistasis_file,
"vennmid",
5,
5,
tight=False,
size=14)
def main(sequence_ratio_file, width, height, pattern, legend):
sequence_ratio = seq_IO.read_sequences(sequence_ratio_file,
additional_params=True)
seqs = [s[0] for s in sequence_ratio]
avg_ratio = [s[1] for s in sequence_ratio]
std = [s[2] for s in sequence_ratio]
label = [s[3] for s in sequence_ratio]
if len(sequence_ratio[0]) > 4:
color = [s[4] for s in sequence_ratio]
else:
color = [convert_label_color(l) for l in label]
#check if std has to be fixed
#if sum([ 1 for a, s in zip(avg_ratio, std) if a - s < 0 ]):
# min_err = [ a - s if a - s >= 0.0 else 0 for a,s in zip(avg_ratio, std) ]
# max_err = [ a + s for a,s in zip(avg_ratio, std) ]
# err = [min_err, max_err]
#else:
# err = std
err = std
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
if legend:
label_legend = [
l if l not in ["CLEAVED", "MIDDLE", "UNCLEAVED"] else None
for l in label
]
patches, labels = bar.draw_actual_plot(axarr[0, 0],
avg_ratio,
color,
"",
"",
"FLAG/HA Ratio",
tick_label=seqs,
yerr=err,
pattern=pattern,
label=label_legend)
lgd = axarr[0, 0].legend(patches,
labels,
loc="upper center",
bbox_to_anchor=(0.5, 1.05),
borderaxespad=0.,
prop={'size': 9},
ncol=2,
fancybox=True)
print patches
print labels
else:
bar.draw_actual_plot(axarr[0, 0],
avg_ratio,
color,
"",
"",
"FLAG/HA Ratio",
tick_label=seqs,
yerr=err,
pattern=pattern)
lgd = None
axarr[0, 0].set_ylim([0, 1.3])
pconv.save_fig(fig,
sequence_ratio_file,
"plot",
width,
height,
tight=True,
size=10,
extra_artists=lgd)
def main(epistasis_file):
dict_epistasis = {
} #list of list of sequences, where each item represents a label
with open(epistasis_file) as e:
lines = e.readlines()
for l in lines[1:]: #ignore header line
tokens = l.split(',')
#value consists of Starting Ratio, Ending Ratio, Epistasis, Ending Fitness, # of Mutations, list of InterSeqs, list of InterFits, list of InterRatios
if dict_epistasis.get((tokens[2], tokens[0])) is None:
dict_epistasis[(tokens[0], tokens[2])] = [
float(tokens[1]),
float(tokens[3]),
float(tokens[5]), tokens[4],
len(tokens[6::3]), tokens[6::3], tokens[7::3], tokens[8::3]
]
neg_epistasis = [0] * 4
no_epistasis = [0] * 4
pos_epistasis = [0] * 4
n_functional = [0] * 4
n_should_be_functional = [0] * 4
n_total = [0] * 4
for i in xrange(2, 6):
ind = i - 2
neg_epistasis[ind] = sum([
1 for key, value in dict_epistasis.items()
if value[2] < -0.000005 and value[4] == i
])
no_epistasis[ind] = sum([
1 for key, value in dict_epistasis.items()
if abs(value[2]) < 0.000005 and value[4] == i
])
pos_epistasis[ind] = sum([
1 for key, value in dict_epistasis.items()
if value[2] > 0.000005 and value[4] == i
])
n_functional[ind] = sum([
1 for key, value in dict_epistasis.items()
if value[3] == "CLEAVED" and value[4] == i
])
n_should_be_functional[ind] = sum([
1 for key, value in dict_epistasis.items()
if all(v == "CLEAVED" for v in value[6]) and value[4] == i
])
n_total[ind] = float(
sum([1 for key, value in dict_epistasis.items() if value[4] == i]))
seq_func = set(
[key[1] for key, val in dict_epistasis.items() if val[3] == "CLEAVED"])
seq_pred_func = set([
key[1] for key, val in dict_epistasis.items()
if all(v == "CLEAVED" for v in val[6])
])
fig, axarr = pconv.create_ax(1, 1, shx=True, shy=True)
fig2, axarr2 = pconv.create_ax(1, 1)
artists = []
artists.extend(
plot_epi(no_epistasis, n_total, axarr[0, 0], "No", color="gray"))
artists.extend(
plot_epi(neg_epistasis,
n_total,
axarr[0, 0],
"Neg.",
bottom=no_epistasis,
color="white"))
artists.extend(
plot_epi(
pos_epistasis,
n_total,
axarr[0, 0],
"Pos.",
bottom=[no + neg for no, neg in zip(no_epistasis, neg_epistasis)],
color="black"))
n_func_frac = [func / total for func, total in zip(n_functional, n_total)]
n_pred_frac = [
pred / total for pred, total in zip(n_should_be_functional, n_total)
]
scatterplot.plot_series(axarr2[0, 0],
[(range(2, 6), n_func_frac, "% Cleaved"),
(range(2, 6), n_pred_frac, "% Pred Cleaved")],
"",
"Number of Mutations",
"Fraction of Total Cases",
size=40,
connect_dots=True,
alpha=1.0)
axarr2[0, 0].set_ylim([0, 4.0])
fig_venn, axarr_venn = pconv.create_ax(1, 1)
venn2([seq_func, seq_pred_func],
set_labels=["Cleaved", "Pred Cleaved"],
ax=axarr_venn[0, 0])
lgd = axarr[0, 0].legend(artists, ["No", "Neg.", "Pos."],
loc="center left",
bbox_to_anchor=(1.05, 0.5),
borderaxespad=0.,
prop={'size': 9},
ncol=1,
fancybox=True)
pconv.save_fig(fig,
epistasis_file,
"plot",
3,
2.5,
tight=False,
size=9,
extra_artists=lgd)
pconv.save_fig(fig2,
epistasis_file,
"pred_v_cl",
5,
5,
tight=True,
size=10)
pconv.save_fig(fig_venn, epistasis_file, "venn", 5, 5, tight=True, size=14)
def main(seq_file, canonical_file, output_prefix):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_file)
print "Beginning Script: {0}".format(datetime.datetime.now())
for canonical in canonical_list_seq:
with open(seq_file) as strings:
seq_list = strings.read().splitlines()
seq_ind_list = [(seq, ind) for ind, seq in enumerate(seq_list)]
orig_len = len(seq_ind_list)
if canonical not in seq_list:
one_away = gsconv.gen_hamdist_one(canonical)
one_away = [o for o in one_away if o != canonical] + [canonical]
seq_ind_list = seq_ind_list[:] + [
(o, ind) for (ind, o) in enumerate(one_away, len(seq_ind_list))
]
edges = [(seq2, seq)
for seq, seq2 in itertools.combinations(seq_ind_list, 2)
if gsconv.hamdist(seq2[0], seq[0]) < 2]
print len(seq_ind_list)
print "Generated Edges: {0}".format(datetime.datetime.now())
numpy.set_printoptions(threshold='nan')
canon_ind = [i for (s, i) in seq_ind_list if s == canonical][0]
T_mat = trans_matrix(seq_ind_list, edges)
#print raise_matrix(T_mat,1)
#print raise_matrix(T_mat,3)
#T = raise_matrix(T_mat,10)
#T = raise_matrix(T_mat,20)
x = [0]
y = [0]
print "Transformed Matrix: {0}".format(datetime.datetime.now())
x.append(1)
y.append(find_frac(T_mat, canon_ind, orig_len))
T_mat_new = T_mat
for i in range(2, 23):
x.append(i)
T_mat_new, frac = square_matrix(T_mat_new, T_mat, canon_ind,
orig_len)
y.append(frac)
print "Raised Matrix {0}: {1}".format(i, datetime.datetime.now())
series.append([x, y, canonical])
fig, ax = conv.create_ax(1, 1)
color = ['orange', 'palevioletred', 'mediumaquamarine', 'deepskyblue']
scatterplot.plot_series(ax[0, 0],
series,
title="",
x_axis="Number of Steps",
colors=color,
y_axis="Fraction Cleaved Variants Reached",
alpha=0.85,
connect_dots=True,
size=15,
edgecolors='k',
linewidth=0)
ax[0, 0].set_xlim(xmin=1)
ax[0, 0].set_ylim(ymin=0.0, ymax=1.0)
ax[0, 0].set_xticks(xrange(1, 23, 3))
lgd = conv.add_legend(ax[0, 0],
location='upper center',
bbox_to_anchor=(0.5, 1.05),
ncol=2,
size=8)
conv.save_fig(fig,
output_prefix,
"fraction_func",
2.5,
3,
size=9.5,
extra_artists=lgd)
print "Outputted Figure: {0}".format(datetime.datetime.now())
def plot_weights_dS(w_dS_sorted, col_dS, unique_z_vals, title1, title2, x_axis_name, y_axis_name, z_axis_name):
n_unique_z_vals = unique_z_vals.shape[0]
fig, axarr = conv.create_ax(n_unique_z_vals, 1, True,True)
#find values in row of total_min
min_vals = w_dS_sorted[w_dS_sorted[:,col_dS]==np.amin(w_dS_sorted[:,col_dS]),:]
min_dS = min_vals[0,col_dS]
max_dS = np.amax(w_dS_sorted[:,col_dS])
dS_list = sorted(w_dS_sorted[:,col_dS].tolist())
nlev = len(dS_list)/10 if len(dS_list)>10 else len(dS_list)/2
levels = dS_list[0::nlev]
subtitle_prefix = z_axis_name + ": "
xlabel = x_axis_name + " Weights"
ylabel = y_axis_name + " Weights"
for ax_ind,z_val in enumerate(sorted(unique_z_vals.tolist())):
ax = axarr[0,ax_ind]
x = w_dS_sorted[w_dS_sorted[:,2]==z_val,0]
y = w_dS_sorted[w_dS_sorted[:,2]==z_val,1]
dS = w_dS_sorted[w_dS_sorted[:,2]==z_val,col_dS]
X = x.reshape(-1, n_unique_z_vals)
Y = y.reshape(-1, n_unique_z_vals)
DS = dS.reshape(-1, n_unique_z_vals)
curr_min_ind = np.argmin(dS)
CS = ax.contourf(X, Y, DS,
extend='both', levels=levels)
CSlines = ax.contour(X, Y, DS, linestyles='solid',
colors=('w',), levels=levels)
if z_val == min_vals[0,2]:
min_x = [min_vals[0,0]]
min_y = [min_vals[0,1]]
min_ds = [min_vals[0,col_dS]]
ann_txt = "Global Minimum: {2:.2f} at ({0:.2f}, {1:.2f})".format(min_x[0], min_y[0], min_ds[0])
ax.scatter(min_x, min_y, c='r', zorder=1)
else:
min_x = [x[curr_min_ind]]
min_y = [y[curr_min_ind]]
min_ds = [dS[curr_min_ind]]
ann_txt = "Minimum: {2:.2f} at ({0:.2f}, {1:.2f})".format(min_x[0], min_y[0], min_ds[0])
ax.scatter(min_x, min_y, c='k', zorder=1)
ax.annotate(ann_txt, xy=(min_x[0], min_y[0]), xytext=(-20,20),
textcoords='offset points', ha='center', va='bottom',
bbox=dict(boxstyle='round,pad=0.2', fc='yellow', alpha=0.3),
arrowprops=dict(arrowstyle='->', connectionstyle='arc3,rad=0.5',
color='red'),size=10)
ax.set_xlabel(xlabel)
ax.set_ylabel(ylabel)
ax.set_title("{0} : {1:.2f}".format(subtitle_prefix,z_val))
cbar = plt.colorbar(CS)
if col_dS == 3:
cbar.ax.set_ylabel('combined_S - Rosetta_S')
else:
cbar.ax.set_ylabel('combined_S - Amber_S')
# Add the contour line levels to the colorbar
cbar.add_lines(CSlines)
filename = args.output_pre + "/" + title1 + "_" + title2 + "_" + str(col_dS) + ".txt"
conv.save_fig(fig, filename, "_weights_vs_deltaS", 4*n_unique_z_vals, 4)
def main(list_sequence_names, canonical_list, output_prefix):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_list)
cleaved_seqs = seq_IO.read_sequences(
[s for s, l in list_sequence_names if l == "CLEAVED"][0])
uncleaved_seqs = seq_IO.read_sequences(
[s for s, l in list_sequence_names if l == "UNCLEAVED"][0])
min_dist = []
avg_dist = []
max_dist = []
for seq in cleaved_seqs:
distances = [conv.hamdist(seq, unc) for unc in uncleaved_seqs]
min_dist.append(min(distances))
avg_dist.append(numpy.mean(distances))
max_dist.append(max(distances))
if seq in canonical_list_seq:
print seq
print min_dist[-1]
print avg_dist[-1]
print max_dist[-1]
fig, ax = pconv.create_ax(1, 3)
hist.draw_actual_plot(ax[0, 0],
min_dist,
"Min. Distance from Boundary",
"Minimum Distances",
log=False,
normed=True,
label=None,
nbins=15,
stacked=False)
hist.draw_actual_plot(ax[1, 0],
avg_dist,
"Avg. Distance from Boundary",
"Average Distances",
log=False,
normed=True,
label=None,
nbins=15,
stacked=False)
hist.draw_actual_plot(ax[2, 0],
max_dist,
"Max. Distance from Boundary",
"Maximum Distances",
log=False,
normed=True,
label=None,
nbins=15,
stacked=False)
#ax[0,0].set_xlim(xmin=1,xmax=5)
#ax[0,0].set_xticks(xrange(1,6))
pconv.save_fig(fig, output_prefix, "dist_from_bounds", 18, 6, size=15)
def main(seq_file, canonical_file, output_prefix):
series = []
canonical_list_seq = seq_IO.read_sequences(canonical_file)
print "Beginning Script: {0}".format(datetime.datetime.now())
for canonical in canonical_list_seq:
with open(seq_file) as strings:
seq_list = strings.read().splitlines()
seq_ind_list = [ (seq, ind) for ind, seq in enumerate(seq_list) ]
orig_len = len(seq_ind_list)
if canonical not in seq_list:
one_away = gsconv.gen_hamdist_one(canonical)
one_away = [ o for o in one_away if o != canonical ] + [canonical]
seq_ind_list = seq_ind_list[:] + [ (o, ind) for (ind, o) in enumerate(one_away, len(seq_ind_list)) ]
edges = [(seq2,seq) for seq,seq2 in itertools.combinations(seq_ind_list,2) if gsconv.hamdist(seq2[0],seq[0]) < 2 ]
print len(seq_ind_list)
print "Generated Edges: {0}".format(datetime.datetime.now())
numpy.set_printoptions(threshold='nan')
canon_ind=[ i for (s, i) in seq_ind_list if s == canonical ][0]
T_mat = trans_matrix(seq_ind_list,edges)
#print raise_matrix(T_mat,1)
#print raise_matrix(T_mat,3)
#T = raise_matrix(T_mat,10)
#T = raise_matrix(T_mat,20)
x = [0]
y = [0]
print "Transformed Matrix: {0}".format(datetime.datetime.now())
x.append(1)
y.append(find_frac(T_mat, canon_ind, orig_len))
T_mat_new = T_mat
for i in range(2,23):
x.append(i)
T_mat_new, frac = square_matrix(T_mat_new,T_mat,canon_ind, orig_len)
y.append(frac)
print "Raised Matrix {0}: {1}".format(i, datetime.datetime.now())
series.append([x,y,canonical])
fig, ax = conv.create_ax(1, 1)
color=['orange', 'palevioletred', 'mediumaquamarine', 'deepskyblue']
scatterplot.plot_series( ax[0,0], series, title="", x_axis="Number of Steps", colors=color, y_axis="Fraction Cleaved Variants Reached", alpha=0.85, connect_dots=True, size=15, edgecolors='k', linewidth=0)
ax[0,0].set_xlim(xmin=1)
ax[0,0].set_ylim(ymin=0.0, ymax=1.0)
ax[0,0].set_xticks(xrange(1,23,3))
lgd = conv.add_legend(ax[0,0], location='upper center', bbox_to_anchor=(0.5, 1.05), ncol=2, size=8)
conv.save_fig(fig, output_prefix, "fraction_func", 2.5, 3, size=9.5, extra_artists=lgd)
print "Outputted Figure: {0}".format(datetime.datetime.now())
def main(list_nodes, output_prefix, metric):
cleaved_seq = {}
uncleaved_seq = {}
middle_seq = {}
for nodes, label in list_nodes:
sequences = seq_IO.read_sequences(nodes, additional_params=True, header=True)
cleaved_seq[label] = { key : val for key, val in sequences.items() if val["type"] == "CLEAVED" }
middle_seq[label] = { key : val for key, val in sequences.items() if val["type"] == "MIDDLE" }
uncleaved_seq[label] = { key : val for key, val in sequences.items() if val["type"] == "UNCLEAVED" }
if metric == "metrics":
labels_non_plot = ["label", "fitness", "type", "canonical"]
orig_labels_to_plot = sorted([ key for key in sequences["DEMEE"].keys() if key not in labels_non_plot ])
labels_to_plot = sorted(orig_labels_to_plot)
else:
orig_labels_to_plot = [metric]
labels_to_plot = [metric]
n_to_plot = len(labels_to_plot)
fig, axarr = pconv.create_ax(n_to_plot, 1, shx=False, shy=False)
nbins = 10
list_seqs = [ k for d in cleaved_seq.values() for k in d.keys() ]
count_seqs = Counter(list_seqs)
#seqs_5_l = [ s for s in list_seqs if count_seqs[s] == 5 ]
seqs_4_l = [ s for s in list_seqs if count_seqs[s] == 4 ]
seqs_3_l = [ s for s in list_seqs if count_seqs[s] == 3 ]
seqs_2_l = [ s for s in list_seqs if count_seqs[s] == 2 ]
seqs_1_l = [ s for s in list_seqs if count_seqs[s] == 1 ]
if metric != "Fraction_Cleaved":
#seqs_5 = list_metrics( cleaved_seq, seqs_5_l, orig_labels_to_plot)
seqs_4 = list_metrics( cleaved_seq, seqs_4_l, orig_labels_to_plot)
seqs_3 = list_metrics( cleaved_seq, seqs_3_l, orig_labels_to_plot)
seqs_2 = list_metrics( cleaved_seq, seqs_2_l, orig_labels_to_plot)
seqs_1 = list_metrics( cleaved_seq, seqs_1_l, orig_labels_to_plot)
for ind, key in enumerate(labels_to_plot):
if key == "pageranks":
log = True
else:
log = False
if key == "Fraction_Cleaved":
data = [ #average_fraction_neighbors_cleaved(cleaved_seq, uncleaved_seq, middle_seq, seqs_5_l),
average_fraction_neighbors_cleaved(cleaved_seq, uncleaved_seq, middle_seq, seqs_4_l),
average_fraction_neighbors_cleaved(cleaved_seq, uncleaved_seq, middle_seq, seqs_3_l),
average_fraction_neighbors_cleaved(cleaved_seq, uncleaved_seq, middle_seq, seqs_2_l),
average_fraction_neighbors_cleaved(cleaved_seq, uncleaved_seq, middle_seq, seqs_1_l)]
normed=True
else:
data = [ #get_data_from_dict(seqs_5, key),
get_data_from_dict(seqs_1, key), get_data_from_dict(seqs_2, key), get_data_from_dict(seqs_3, key), get_data_from_dict(seqs_4, key) ]
normed=True
hist.draw_actual_plot(axarr[0,ind], data, "", key.capitalize(), colors = [ tuple(c) for c in plt.cm.Blues(np.linspace(0.2, 1, 4)).tolist()], log=log, normed=normed, label=["Cl. by 5", "Cl. by 4", "Cl. by 3", "Cl. by 2", "Cl. by 1"], nbins=nbins)
axarr[0,ind].ticklabel_format(axis='x', style='sci', scilimits=(-2,2))
#pconv.add_legend(axarr[0,ind], location="upper right")
pconv.save_fig(fig, output_prefix, metric, n_to_plot*3, 3, tight=True, size=9)
fig_bar, axarr_bar = pconv.create_ax(1, 1, shx=False, shy=False)
gradient = np.linspace(1, 0.2, 256)
#gradient = np.hstack((gradient, gradient))
gradient = np.array(zip(gradient,gradient))
axarr_bar[0,0].imshow(gradient, aspect='auto', cmap=plt.get_cmap('Blues'))
#axarr_bar[0,0].set_axis_off()
plt.tick_params(
axis='both', # changes apply to the x-axis
which='both', # both major and minor ticks are affected
bottom='off', # ticks along the bottom edge are off
top='off', # ticks along the top edge are off
labelbottom='off', # labels along the bottom edge are off
left='off', # ticks along the bottom edge are off
right='off', # ticks along the top edge are off
labelright='off') # labels along the bottom edge are off
pconv.save_fig(fig_bar, output_prefix, "colorbar", 0.3, 3, tight=True)
def main(seq_file, canonical_file, output_prefix):
#canonical_list_seq = seq_IO.read_sequences(canonical_file)
canonical_list_seq = ["DEMEE","DEMED"]
print "Beginning Script: {0}".format(datetime.datetime.now())
with open(seq_file) as strings:
seq_list = strings.read().splitlines()
seq_ind_list = [ (seq, ind) for ind, seq in enumerate(seq_list) ]
seq_ind_dict = { seq : ind for seq, ind in seq_ind_list }
orig_len = len(seq_ind_list)
edges = []
edges_set = set()
print "Read in Data: {0}".format(datetime.datetime.now())
for seq, seq_ind in seq_ind_dict.items():
neighbors = gsconv.gen_hamdist_one(seq)
edges_set.update([ (seq, n) for n in neighbors if n in seq_ind_dict ])
edges += [((seq, seq_ind), (n,seq_ind_dict[n])) for n in neighbors if n in seq_ind_dict and (n,seq) not in edges_set ]
print len(seq_ind_list)
print "Generated Edges: {0}".format(datetime.datetime.now())
numpy.set_printoptions(threshold='nan')
canon_ind_dict = { canonical : [ i for (s, i) in seq_ind_list if s == canonical ][0] for canonical in canonical_list_seq }
T_mat = trans_matrix(seq_ind_list,edges)
#print raise_matrix(T_mat,1)
#print raise_matrix(T_mat,3)
#T = raise_matrix(T_mat,10)
#T = raise_matrix(T_mat,20)
print "Transformed Matrix: {0}".format(datetime.datetime.now())
canon_x = { can : [0,1] for can in canonical_list_seq }
canon_y = { can : [0.0, find_frac(T_mat, canon_ind_dict[can], orig_len)] for can in canonical_list_seq }
print "Made x and y dicts: {0}".format(datetime.datetime.now())
T_mat_new = T_mat
for i in range(2,23):
T_mat_new = square_matrix(T_mat_new, T_mat)
for can in canonical_list_seq:
canon_x[can].append(i)
canon_y[can].append(find_frac(T_mat_new, canon_ind_dict[can], orig_len))
print "Raised Matrix {0}: {1}".format(i, datetime.datetime.now())
series = [ [canon_x[can],canon_y[can], can] for can in canonical_list_seq ]
fig, ax = conv.create_ax(1, 1)
color=['orange', 'palevioletred', 'mediumaquamarine', 'deepskyblue']
scatterplot.plot_series( ax[0,0], series, title="", x_axis="Number of Steps", colors=color, y_axis="Fraction Cleaved Variants Reached", alpha=0.85, connect_dots=True, size=15, edgecolors='k', linewidth=0)
ax[0,0].set_xlim(xmin=1)
ax[0,0].set_ylim(ymin=0.0, ymax=1.0)
ax[0,0].set_xticks(xrange(1,23,3))
lgd = conv.add_legend(ax[0,0], location='upper center', bbox_to_anchor=(0.5, 1.05), ncol=2, size=8)
conv.save_fig(fig, output_prefix, "fraction_func", 2.5, 3, size=9.5, extra_artists=lgd)
print "Outputted Figure: {0}".format(datetime.datetime.now())
def main(seq_file, canonical_file, output_prefix):
#canonical_list_seq = seq_IO.read_sequences(canonical_file)
canonical_list_seq = ["DEMEE", "DEMED"]
print "Beginning Script: {0}".format(datetime.datetime.now())
with open(seq_file) as strings:
seq_list = strings.read().splitlines()
seq_ind_list = [(seq, ind) for ind, seq in enumerate(seq_list)]
seq_ind_dict = {seq: ind for seq, ind in seq_ind_list}
orig_len = len(seq_ind_list)
edges = []
edges_set = set()
print "Read in Data: {0}".format(datetime.datetime.now())
for seq, seq_ind in seq_ind_dict.items():
neighbors = gsconv.gen_hamdist_one(seq)
edges_set.update([(seq, n) for n in neighbors if n in seq_ind_dict])
edges += [((seq, seq_ind), (n, seq_ind_dict[n])) for n in neighbors
if n in seq_ind_dict and (n, seq) not in edges_set]
print len(seq_ind_list)
print "Generated Edges: {0}".format(datetime.datetime.now())
numpy.set_printoptions(threshold='nan')
canon_ind_dict = {
canonical: [i for (s, i) in seq_ind_list if s == canonical][0]
for canonical in canonical_list_seq
}
T_mat = trans_matrix(seq_ind_list, edges)
#print raise_matrix(T_mat,1)
#print raise_matrix(T_mat,3)
#T = raise_matrix(T_mat,10)
#T = raise_matrix(T_mat,20)
print "Transformed Matrix: {0}".format(datetime.datetime.now())
canon_x = {can: [0, 1] for can in canonical_list_seq}
canon_y = {
can: [0.0, find_frac(T_mat, canon_ind_dict[can], orig_len)]
for can in canonical_list_seq
}
print "Made x and y dicts: {0}".format(datetime.datetime.now())
T_mat_new = T_mat
for i in range(2, 23):
T_mat_new = square_matrix(T_mat_new, T_mat)
for can in canonical_list_seq:
canon_x[can].append(i)
canon_y[can].append(
find_frac(T_mat_new, canon_ind_dict[can], orig_len))
print "Raised Matrix {0}: {1}".format(i, datetime.datetime.now())
series = [[canon_x[can], canon_y[can], can] for can in canonical_list_seq]
fig, ax = conv.create_ax(1, 1)
color = ['orange', 'palevioletred', 'mediumaquamarine', 'deepskyblue']
scatterplot.plot_series(ax[0, 0],
series,
title="",
x_axis="Number of Steps",
colors=color,
y_axis="Fraction Cleaved Variants Reached",
alpha=0.85,
connect_dots=True,
size=15,
edgecolors='k',
linewidth=0)
ax[0, 0].set_xlim(xmin=1)
ax[0, 0].set_ylim(ymin=0.0, ymax=1.0)
ax[0, 0].set_xticks(xrange(1, 23, 3))
lgd = conv.add_legend(ax[0, 0],
location='upper center',
bbox_to_anchor=(0.5, 1.05),
ncol=2,
size=8)
conv.save_fig(fig,
output_prefix,
"fraction_func",
2.5,
3,
size=9.5,
extra_artists=lgd)
print "Outputted Figure: {0}".format(datetime.datetime.now())
def main(rec_corr_path, ddg_path, amber_pdb_path, rosetta_pdb_path, out_csv_path):
amber_csv_path = os.path.join(ddg_path, "amber")
rosetta_csv_path = os.path.join(ddg_path, "rosetta")
amber_inter_mean_path = os.path.join(ddg_path, "amber_inter_mean")
rosetta_inter_mean_path = os.path.join(ddg_path, "rosetta_inter_mean")
rosetta_inter_path = os.path.join(ddg_path, "rosetta_inter")
#Plots to generate
#Per pdb - known ddg vs. many diff protocols, each with their own row. A protocol may have more than one plot depending on filtering method (i.e. mean, bottom 3, pareto)
#For all pdbs - pred rosetta corr values vs. many diff protocols, each with their series color. A protocol may have more than one plot depending on filtering method (i.e. mean, bottom 3, pareto). 3 rows one for each corr value
#For all pdbs - known ddg vs. many diff protocols, each with their own row. A protocol may have more than one plot depending on filtering method (i.e. mean, bottom 3, pareto)
list_rec_corr_names = glob.glob(rec_corr_path + "*.rc")
#corr_values_dict has the following shape - "Pred" : { "Pred" : [ddg_vals] }, "Amber" : { "Mean.." : [ddg_vals], "Bott.." : [ddg_vals]}, "Rosetta" : { "Mean.." : [ddg_vals], "Bott.." : [ddg_vals]}
corr_values_dict = {}
all_amber_ddg_dict = {}
all_rosetta_ddg_dict = {}
all_known_ddg_dict = {}
all_pred_ddg_dict = {}
k_ddg = []
p_ddg = []
for rec_corr in list_rec_corr_names:
print rec_corr
rec_corr_list = read_csv_list(rec_corr)
#no known ddg
if len(rec_corr_list[0]) == 3:
continue
amber_dg_dict = {}
rosetta_dg_dict = {}
#read in all amber csvs that correspond to column 3 in rec_corr file and rosetta ones too
for record_id, prefix, filename, known_ddg, pred_ddg in rec_corr_list:
amber_dg_dict[filename] = { "Mean Binding Energy" : get_mean_csv(os.path.join(amber_csv_path,filename+".csv"), protocol="amber"),
"Bottom 3 Binding Energy" : get_bottom3_csv(os.path.join(amber_csv_path,filename+".csv"), protocol="amber"),
"Mean Interaction Energy" : get_mean_txt(os.path.join(amber_inter_mean_path,filename+".txt")) }
rosetta_dg_dict[filename] = { "Mean Binding Energy" : get_mean_csv(os.path.join(rosetta_csv_path,filename+".csv"), protocol="rosetta"),
"Bottom 3 Binding Energy" : get_bottom3_csv(os.path.join(rosetta_csv_path,filename+".csv"), protocol="rosetta"),
"Mean Interaction Energy" : get_mean_txt(os.path.join(rosetta_inter_mean_path,filename+".txt")),
"Bottom 3 Interaction Energy" : get_bottom3_csv(os.path.join(rosetta_inter_path,filename+".csv")) }
#find wt csv that correspond to wt row in rec_corr_file (column 2)
wt_csv_name = [ rec[2] for rec in rec_corr_list if "wt" in rec[1] ][0]
amber_ddg_dict = {}
rosetta_ddg_dict = {}
known_ddg_dict = {}
pred_ddg_dict = {}
#loops thru other records in rec_corr_dict
for rec, prefix, filename, k,p in rec_corr_list:
if "wt" not in prefix:
if amber_ddg_dict.get(filename) is None:
amber_ddg_dict[filename] = {}
if rosetta_ddg_dict.get(filename) is None:
rosetta_ddg_dict[filename] = {}
for key, dg in amber_dg_dict[wt_csv_name].items():
amber_ddg_dict[filename][key] = amber_dg_dict[filename][key] - dg
for key, dg in rosetta_dg_dict[wt_csv_name].items():
rosetta_ddg_dict[filename][key] = rosetta_dg_dict[filename][key] - dg
known_ddg_dict[filename] = { "Known" : float(k) }
pred_ddg_dict[filename] = { "Pred" : float(p) }
all_amber_ddg_dict.update(amber_ddg_dict)
all_rosetta_ddg_dict.update(rosetta_ddg_dict)
all_known_ddg_dict.update(known_ddg_dict)
all_pred_ddg_dict.update(pred_ddg_dict)
fig, axarr = conv.create_ax(max([len(d) for k, d in amber_dg_dict.items() ]+[len(d) for k,d in rosetta_dg_dict.items()]), 3, shx=True, shy=True)
plot_ddg_dict(rosetta_ddg_dict,known_ddg_dict,axarr,0,"Rosetta",corr_values_dict)
plot_ddg_dict(amber_ddg_dict,known_ddg_dict,axarr,1,"Amber",corr_values_dict)
plot_ddg_dict(pred_ddg_dict,known_ddg_dict,axarr,2,"Pred",corr_values_dict)
conv.save_fig(fig, out_csv_path + "/" + os.path.splitext(os.path.basename(rec_corr))[0] + ".txt", "ddg", max([len(d) for k, d in amber_dg_dict.items() ]+[len(d) for k,d in rosetta_dg_dict.items()])*4, 12)
#Plot all correlation values
fig_all, axarr_all = conv.create_ax(len(corr_values_dict["Rosetta"]),3)
#assumes that Rosetta has more protocols than Amber
for x_ind,(protocol, vals) in enumerate(corr_values_dict["Rosetta"].items()):
if corr_values_dict["Amber"].get(protocol) is not None:
amber_vals = corr_values_dict["Amber"][protocol]
else:
amber_vals = None
pred_vals = corr_values_dict["Pred"]["Pred"]
labels=["-PCC","-Rho","-Mae"]
for ind,(val_list,label) in enumerate(zip(vals,labels)):
series = [[val_list,pred_vals[ind],"Rosetta "+protocol]]
if amber_vals is not None:
series.append([amber_vals[ind],pred_vals[ind],"Amber "+protocol])
scatterplot.plot_series(axarr_all[ind,x_ind], series, protocol,"Pred",label,colors=['coral','cyan'], size=40)
scatterplot.add_x_y_line(axarr_all[ind,x_ind])
#if x_ind == 2:
# axarr_all[x_ind,y_ind].set_xlim([-0.2,10.0])
# axarr_all[x_ind,y_ind].set_ylim([-0.2,10.0])
# scatterplot.add_x_y_line(axarr_all[x_ind,y_ind],0.0,10.0)
#else:
#axarr_all[x_ind,y_ind].set_xlim([-1.2,1.2])
# axarr_all[x_ind,y_ind].set_ylim([-1.2,1.2])
# scatterplot.add_x_y_line(axarr_all[x_ind,y_ind],-1.0,1.0)
conv.save_fig(fig_all, out_csv_path + "/all.txt", "ddg", 16, 12)
fig_all_corr, axarr_all_corr = conv.create_ax(max([len(d) for k, d in all_amber_ddg_dict.items() ]+[len(d) for k,d in all_rosetta_ddg_dict.items()]), 3, shx=True, shy=True)
plot_ddg_dict(all_rosetta_ddg_dict,all_known_ddg_dict,axarr_all_corr,0,"Rosetta",corr_values_dict)
plot_ddg_dict(all_amber_ddg_dict,all_known_ddg_dict,axarr_all_corr,1,"Amber",corr_values_dict)
plot_ddg_dict(all_pred_ddg_dict,all_known_ddg_dict,axarr_all_corr,2,"Pred",corr_values_dict)
conv.save_fig(fig_all_corr, out_csv_path + "/all_corr.txt", "ddg",max([len(d) for k, d in amber_dg_dict.items() ]+[len(d) for k,d in rosetta_dg_dict.items()])*4, 12)
def main(input_dir_1, scoretype1, input_dir_2, scoretype2, rmsd_cutoff, output_pre ):
#read in and rename arguments
title1 = os.path.basename(input_dir_1)
title2 = os.path.basename(input_dir_2)
d1, n1 = scorefileparse.read_dec_nat(input_dir_1, scoretype1, repl_orig=False)
d2, n2 = scorefileparse.read_dec_nat(input_dir_2, scoretype2, repl_orig=False)
dec1 = scorefileparse.filter_pdbs_by_rmsd(d1, rmsd_cutoff)
nat1 = scorefileparse.filter_pdbs_by_rmsd(n1, rmsd_cutoff)
dec2 = scorefileparse.filter_pdbs_by_rmsd(d2, rmsd_cutoff)
nat2 = scorefileparse.filter_pdbs_by_rmsd(n2, rmsd_cutoff)
dec_norm1 = scorefileparse.norm_pdbs(dec1)
nat_norm1 = scorefileparse.norm_pdbs(nat1,dec1)
dec_norm2 = scorefileparse.norm_pdbs(dec2)
nat_norm2 = scorefileparse.norm_pdbs(nat2,dec2)
[dec_inter1, nat_inter1, dec_inter2, nat_inter2] = scorefileparse.pdbs_intersect([dec_norm1, nat_norm1, dec_norm2, nat_norm2])
[dec_inter1, dec_inter2] = scorefileparse.pdbs_scores_intersect([dec_inter1, dec_inter2])
[nat_inter1, nat_inter2] = scorefileparse.pdbs_scores_intersect([nat_inter1, nat_inter2])
dec_filt1 = scorefileparse.filter_norm_pdbs(dec_norm1)
nat_filt1 = scorefileparse.filter_norm_pdbs(nat_norm1)
dec_filt2 = scorefileparse.filter_norm_pdbs(dec_norm2)
nat_filt2 = scorefileparse.filter_norm_pdbs(nat_norm2)
[dec_finter1, dec_finter2] = scorefileparse.pdbs_scores_intersect([dec_filt1, dec_filt2])
[nat_finter1, nat_finter2] = scorefileparse.pdbs_scores_intersect([nat_filt1, nat_filt2])
fig, axarr = conv.create_ax(2, len(dec_inter1))
line_plot_data = {}
min_naive_by_pdb = {}
for x_ind,pdb in enumerate(sorted(dec_inter1.keys())):
ax = axarr[x_ind, 0]
plot_r_v_r(dec_inter1, dec_inter2, nat_inter1, nat_inter2, ax, pdb, title1, title2)
ax = axarr[x_ind, 1]
min_naive = plot_pareto(dec_inter1, dec_inter2, nat_inter1, nat_inter2, ax, pdb, title1, title2)
keys_to_include = ["Amber", "Rosetta","All","Pareto10"]
for key, (rank1, rank2, rmsd) in min_naive.items():
#if key not in keys_to_include:
# continue
if line_plot_data.get(key) is None:
line_plot_data[key] = ([],[])
line_plot_data[key][0].append(pdb)
line_plot_data[key][1].append(rmsd)
if min_naive_by_pdb.get(pdb) is None:
min_naive_by_pdb[pdb] = {}
min_naive_by_pdb[pdb][key] = rmsd
#organize data
indices = list(range(len(line_plot_data["All"][1])))
indices.sort(key=lambda x: line_plot_data["All"][1][x])
ranked_pdbs_by_rmsd_all = {}
for i, x in enumerate(indices):
ranked_pdbs_by_rmsd_all[line_plot_data["All"][0][x]] = i
for label, (pdbs, rmsds) in line_plot_data.items():
line_plot_data[label] = tuple(zip(*sorted(zip(pdbs,rmsds), key=lambda x: ranked_pdbs_by_rmsd_all[x[0]] )))
filename = output_pre + "/" + title1 + "_" + title2 + ".txt"
#suffix="rmsd_v_rmsd_{0}".format(rmsd_cutoff)
#conv.save_fig(fig, filename, suffix, 7, len(dec_inter1)*3)
#plot line plot
all_pareto_labels = []
for initial in ["R","A"]:
ordered_labels = ["All", "Amber", "Rosetta"]
for i in range(1,11):
ordered_labels.append("Pareto{0}{1}".format(initial,i))
all_pareto_labels.append("Pareto{0}{1}".format(initial,i))
lines = [ (line_plot_data[label][0], line_plot_data[label][1], label) for label in ordered_labels ]
fig2, axarr2 = conv.create_ax(1, len(ordered_labels), shx=True, shy=True)
for i, label in enumerate(ordered_labels):
line.plot_series(axarr2[i,0], lines[0:i+1], "RMSD vs. pdb", "PDB", "RMSD", linestyle='')
conv.add_legend(axarr2[i,0])
conv.save_fig(fig2, filename, "_line_{0}".format(initial), 10, len(ordered_labels)*5)
#plot histogram plot
hist_comp = [ ("Amber","All"), ("Rosetta", "All"), ("ParetoR10", "All"), ("ParetoA10", "All")]
hist_comp.extend([ ("ParetoR{0}".format(ind),"Rosetta") for ind in range(1,11) ])
hist_comp.extend([ ("ParetoR{0}".format(ind),"Amber") for ind in range(1,11) ])
hist_comp.extend([ ("ParetoA{0}".format(ind),"Rosetta") for ind in range(1,11) ])
hist_comp.extend([ ("ParetoA{0}".format(ind), "Amber") for ind in range(1,11) ])
fig3, axarr3 = conv.create_ax(2, len(hist_comp), shx=False, shy=False)
for ind, (top, bottom) in enumerate(hist_comp):
gen_dist_plot(axarr3[ind,0], axarr3[ind,1], top, bottom, min_naive_by_pdb)
conv.save_fig(fig3, filename, "_distdeltas", 7, len(hist_comp)*5, tight=False)
#plot scatterplot
fig4, axarr4 = conv.create_ax(10, 2)
for i in range(1,11):
gen_scatterplot(axarr4[0,i-1], "ParetoR{0}".format(i), "Rosetta", "Amber", min_naive_by_pdb)
gen_scatterplot(axarr4[1,i-1], "ParetoA{0}".format(i), "Rosetta", "Amber", min_naive_by_pdb)
conv.save_fig(fig4, filename, "_scattdeltas", 30, 6)
def main(list_nodes, output_prefix, metric, create_keys=False):
if not create_keys:
sequences = seq_IO.read_sequences(list_nodes,
additional_params=True,
header=True)
else:
sequences = seq_IO.read_sequences(list_nodes,
additional_params=True,
header=True,
create_keys=True)
cleaved_seq = {
key: val
for key, val in sequences.items() if val["type"] == "CLEAVED"
}
middle_seq = {
key: val
for key, val in sequences.items() if val["type"] == "MIDDLE"
}
uncleaved_seq = {
key: val
for key, val in sequences.items() if val["type"] == "UNCLEAVED"
}
print len(cleaved_seq)
if metric == "metrics":
labels_non_plot = ["label", "fitness", "type", "canonical", "timeset"]
#labels_to_plot = sorted([ key for key in sequences["YNYIN"].keys() if key not in labels_non_plot ] + ["Fraction_Cleaved"])
labels_to_plot = sorted([
key for key in sequences["YNYIN"].keys()
if key not in labels_non_plot
])
else:
labels_to_plot = [metric]
n_to_plot = len(labels_to_plot)
fig, axarr = pconv.create_ax(n_to_plot, 1, shx=False, shy=False)
nbins = 10
for ind, key in enumerate(labels_to_plot):
if key == "pageranks":
log = True
else:
log = False
if key == "Fraction_Cleaved":
# data = [ conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), cleaved_seq.keys()).values(),
# conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), middle_seq.keys()).values(),
# conv.fraction_neighbors_cleaved(cleaved_seq.keys(), uncleaved_seq.keys(), middle_seq.keys(), uncleaved_seq.keys()).values()]
normed = True
else:
data = [
get_data_from_dict(cleaved_seq, key),
get_data_from_dict(middle_seq, key),
get_data_from_dict(uncleaved_seq, key)
]
normed = True
print key
hist.draw_actual_plot(axarr[0, ind],
data,
"",
key.capitalize(),
log=log,
normed=normed,
label=["Cleaved", "Middle", "Uncleaved"],
nbins=nbins)
axarr[0, ind].ticklabel_format(axis='x',
style='sci',
scilimits=(-2, 2))
#pconv.add_legend(axarr[0,ind], location="middle right")
pconv.save_fig(fig,
output_prefix,
metric,
n_to_plot * 2.5,
2.5,
tight=True,
size=9)
def main(list_input_dirs, energies_names, output_pre):
#read in and rename arguments
inp_dir1=list_input_dirs[0][0]
scoretype1=list_input_dirs[0][1]
inp_dir2=list_input_dirs[1][0]
scoretype2=list_input_dirs[1][1]
title1 = os.path.basename(inp_dir1)
title2 = os.path.basename(inp_dir2)
column_dict = {}
for c in energies_names:
column_dict[c[0]] = c[1:]
dec1, nat1 = scorefileparse.read_dec_nat(inp_dir1, energies_names[scoretype1], scoretype1)
dec2, nat2 = scorefileparse.read_dec_nat(inp_dir2, energies_names[scoretype2], scoretype2)
[dec_inter1, nat_inter1, dec_inter2, nat_inter2] = scorefileparse.pdbs_intersect([dec1, nat1, dec2, nat2])
sum_discs = Counter()
fig, axarr = conv.create_ax(1, len(dec_inter1)+1, True,True)
for x_ind, pdb in enumerate(sorted(dec_inter1.keys())):
discs_per_pdb = {}
for w_1 in xrange(-10,10,2):
for w_2 in xrange(-10,10,2):
weight_1 = 2 ** w_1
weight_2 = 2 ** w_2
weighted_1 = scorefileparse.weight_dict(dec_inter1[pdb], weight_1)
weighted_2 = scorefileparse.weight_dict(dec_inter2[pdb], weight_2)
merged = scorefileparse.merge_dicts([weighted_1, weighted_2])
ddata1 = scorefileparse.convert_disc(merged)
disc_divs = [1.0,1.5,2.0,2.5,3.0,4.0,6.0]
disc1, d, counts = disc.given_data_run_disc(ddata1, True, disc_divs)
discs_per_pdb[(weight_1,weight_2)] = disc1
sorted_disc = sorted(discs_per_pdb.values())
max_title = [ t for t,v in discs_per_pdb.items() if v == sorted_disc[0] ]
#header_string = "\t".join("{0:.3f}-{1:.3f}".format(x,y) for x,y in sorted(discs_per_pdb.keys())) + "\tMax_Weight"
#values_string = "\t".join(format(x, "10.3f") for (w1,w2),x in sorted(discs_per_pdb.items())) + "\t{0:.3f}".format(max_title[0])
#print header_string
#print values_string
ax = axarr[x_ind, 0]
#ax.set_xlim(-10, 600)
#ax.set_ylim(-10, 600)
ax.set_xscale('log', basex=2)
ax.set_yscale('log', basey=2)
x = [ w1 for (w1,w2) in sorted(discs_per_pdb.keys()) ]
y = [ w2 for (w1,w2) in sorted(discs_per_pdb.keys()) ]
d = [ v for k,v in sorted(discs_per_pdb.items()) ]
min_y = min(discs_per_pdb.values())
max_y = max(discs_per_pdb.values())
#print min_y, max_y
s = scatterplot.draw_actual_plot(ax, x, y, d, pdb, scoretype1, scoretype2, 'bwr')
fig.colorbar(s,ax=ax)
#ax.axhline(y=min_y)
#ax.set_ylim(min_y-0.05,max_y+0.05)
scatterplot.add_x_y_line(ax, 0,600)
sum_discs.update(discs_per_pdb)
#print "All PDBs {0}".format(len(dec_inter1))
#sorted_disc = sorted(sum_discs.values())
#max_title = [ t for t,v in sum_discs.items() if v == sorted_disc[0] ]
#header_string = "\t".join(format(x, "10.3f") for x in sorted(sum_discs.keys())) + "\tMax_Weight"
#values_string = "\t".join(format(x/len(dec_inter1), "10.3f") for key,x in sorted(sum_discs.items())) + "\t{0:.3f}".format(max_title[0])
#print header_string
#print values_string
ax = axarr[len(dec_inter1), 0]
min_y = min(x/len(dec_inter1) for x in sum_discs.values())
max_y = max(x/len(dec_inter1) for x in sum_discs.values())
x = [ w1 for w1,w2 in sorted(sum_discs.keys()) ]
y = [ w2 for w1,w2 in sorted(sum_discs.keys()) ]
d = [ v/len(dec_inter1) for k,v in sorted(sum_discs.items()) ]
#fix titles of axes
ax.set_xscale('log', basex=2)
ax.set_yscale('log', basey=2)
s = scatterplot.draw_actual_plot(ax, x,y,d, "All", scoretype1, scoretype2, cm='bwr')
fig.colorbar(s,ax=ax)
scatterplot.add_x_y_line(ax, 0,600)
#ax.axhline(y=min_y)
conv.save_fig(fig, output_pre, "_weights_v_disc", 3, len(dec_inter1)*3)