def from_paths(cls,
analysis_paths_and_ids,
glycopeptide_hypothesis_path,
glycopeptide_hypothesis_id,
glycan_hypothesis_path,
glycan_hypothesis_id,
unobserved_penalty_scale=None,
lambda_limit=0.2,
require_multiple_observations=True,
observation_aggregator=None,
output_path=None,
n_threads=4,
q_value_threshold=0.05):
gp_db = GlycopeptideDiskBackedStructureDatabase(
glycopeptide_hypothesis_path, glycopeptide_hypothesis_id)
gc_db = GlycanCompositionDiskBackedStructureDatabase(
glycan_hypothesis_path, glycan_hypothesis_id)
analyses = [
serialize.AnalysisDeserializer(conn, analysis_id=an_id)
for conn, an_id in analysis_paths_and_ids
]
inst = cls(analyses,
gp_db,
gc_db,
unobserved_penalty_scale=unobserved_penalty_scale,
lambda_limit=lambda_limit,
require_multiple_observations=require_multiple_observations,
observation_aggregator=observation_aggregator,
output_path=output_path,
n_threads=n_threads,
q_value_threshold=q_value_threshold)
return inst
def make_template_stream(self):
template_obj = self.env.get_template("overview.templ")
ads = serialize.AnalysisDeserializer(
self.database_connection._original_connection,
analysis_id=self.analysis_id)
hypothesis = ads.analysis.hypothesis
sample_run = ads.analysis.sample_run
template_stream = template_obj.stream(
analysis=ads.analysis,
hypothesis=hypothesis,
sample_run=sample_run,
protein_index=self.protein_index,
glycoprotein_iterator=self.iterglycoproteins(),
renderer=self,
)
return template_stream
def make_template_stream(self):
template_obj = self.env.get_template("overview.templ")
ads = serialize.AnalysisDeserializer(
self.database_connection._original_connection,
analysis_id=self.analysis_id)
hypothesis = ads.analysis.hypothesis
sample_run = ads.analysis.sample_run
if self.use_dynamic_display_mode:
self.status_update("Using dynamic display mode")
template_stream = template_obj.stream(
analysis=ads.analysis,
hypothesis=hypothesis,
sample_run=sample_run,
protein_index=self.protein_index,
glycoprotein_iterator=self.iterglycoproteins(),
renderer=self,
use_dynamic_display_mode=self.use_dynamic_display_mode)
return template_stream
def make_template_stream(self):
template_obj = self.env.get_template("overview.templ")
ads = serialize.AnalysisDeserializer(
self.database_connection._original_connection,
analysis_id=self.analysis_id)
self.glycan_chromatograms = gcs = ads.load_glycan_composition_chromatograms(
)
# und = ads.load_unidentified_chromatograms()
self.unidentified_chromatograms = und = ChromatogramFilter(
ads.query(serialize.UnidentifiedChromatogram).filter(
serialize.UnidentifiedChromatogram.analysis_id ==
self.analysis_id).all())
if len(gcs) == 0:
self.log(
"No glycan compositions were identified. Skipping report building"
)
templ = Template('''
<html>
<style>
body {
font-family: sans-serif;
}
</style>
<body>
<h3>No glycan compositions were identified</h3>
</body>
</html>
''')
return templ.stream()
summary_plot = summaries.GlycanChromatographySummaryGraphBuilder(
filter(lambda x: x.score > self.threshold, gcs + und))
lcms_plot, composition_abundance_plot = summary_plot.draw(min_score=5)
try:
lcms_plot.ax.legend_.set_visible(False)
except AttributeError:
# The legend may not have been created
pass
lcms_plot.ax.set_title("Glycan Composition\nLC-MS Aggregated EICs",
fontsize=24)
fig = lcms_plot.ax.figure
fig.set_figwidth(fig.get_figwidth() * 2.)
fig.set_figheight(fig.get_figheight() * 2.)
composition_abundance_plot.ax.set_title(
"Glycan Composition\nTotal Abundances", fontsize=24)
composition_abundance_plot.ax.set_xlabel(
composition_abundance_plot.ax.get_xlabel(), fontsize=14)
def resolve_key(key):
match = gcs.find_key(key)
if match is None:
match = und.find_key(key)
return match
template_stream = (template_obj.stream(
analysis=ads.analysis,
lcms_plot=svguri_plot(lcms_plot.ax,
bbox_inches='tight',
patchless=True,
svg_width="100%"),
composition_abundance_plot=svguri_plot(
composition_abundance_plot.ax,
bbox_inches='tight',
patchless=True,
svg_width="100%"),
glycan_chromatograms=gcs,
unidentified_chromatograms=und,
resolve_key=resolve_key))
return template_stream
import glypy
from glypy.composition.composition_transform import strip_derivatization
import numpy as np
import pandas as pd
from sklearn.metrics import auc as sk_auc, roc_curve, roc_auc_score, precision_recall_curve
from matplotlib import pyplot as plt, rcParams
rcParams['figure.figsize'] = 8, 6
Formate = chromatogram_tree.Formate
Ammonium = chromatogram_tree.Ammonium
Unmodified = chromatogram_tree.Unmodified
print "Native Plots"
native_path = "analysis/results/sulfated/phil-82-native-fit-prior.db"
ads = serialize.AnalysisDeserializer(native_path)
und = trace.ChromatogramFilter(
ads.query(serialize.UnidentifiedChromatogram).all())
gcs = trace.ChromatogramFilter(
ads.query(serialize.GlycanCompositionChromatogram).all())
print len(filter(lambda x: x.score > 5 and not x.used_as_adduct, gcs + und))
summary_plot = summaries.GlycanChromatographySummaryGraphBuilder(
filter(lambda x: x.score > 5 and not x.used_as_adduct, gcs + und))
lcms_plot, composition_abundance_plot = summary_plot.draw(min_score=5)
lcms_plot.ax.legend_.set_visible(False)
lcms_plot.ax.set_xlabel("Retention Time (Min)", fontsize=14)
lcms_plot.ax.set_ylabel("Relative Abundance", fontsize=14)
from matplotlib import pyplot as plt, rcParams
rcParams['figure.figsize'] = 8, 6
Formate = chromatogram_tree.Formate
Ammonium = chromatogram_tree.Ammonium
Unmodified = chromatogram_tree.Unmodified
files = {
"combinatorial_unregularized":
"./analysis/results/igg-native-unregularized.db",
"combinatorial_partial": "./analysis/results/igg-native-fit-prior.db",
"combinatorial_grid": "./analysis/results/igg-native-grid.db",
}
deserializers = {
key: serialize.AnalysisDeserializer(value)
for key, value in files.items()
}
print("Loading Datasets")
gcs_map = {
key: trace.ChromatogramFilter(
value.query(serialize.GlycanCompositionChromatogram))
for key, value in deserializers.items()
}
all_keys = set()
for group, members in gcs_map.items():
for member in members:
all_keys.add(member.key)
true_positives = set(
import numpy as np
import pandas as pd
from sklearn.metrics import auc as sk_auc, roc_curve, roc_auc_score, precision_recall_curve
from matplotlib import pyplot as plt, rcParams
rcParams['figure.figsize'] = 8, 6
Formate = chromatogram_tree.Formate
Ammonium = chromatogram_tree.Ammonium
Unmodified = chromatogram_tree.Unmodified
import variable_writer
db = database.GlycanCompositionDiskBackedStructureDatabase("./analysis/hypothesis/reduced-permethylated.db", 1)
ads = serialize.AnalysisDeserializer("./analysis/results/serum-rp-unregularized.db")
ads2 = serialize.AnalysisDeserializer("./analysis/results/serum-rp-prior.db")
ads3 = serialize.AnalysisDeserializer("./analysis/results/serum-rp-grid.db")
with open("./analysis/multiglycan-match.csv", "rb") as f:
obs = list(csv.reader(f))
true_positives = []
false_positives = []
for row in obs:
gc = glypy.glycan_composition.HashableGlycanComposition.parse(row[0])
status = row[1]
try:
remark = row[2]
except IndexError:
db = database.GlycanCompositionDiskBackedStructureDatabase("./analysis/hypothesis/native.db", 1)
files = {
"combinatorial_unregularized": "./analysis/results/sulfated/phil-bs-native-unregularized.db",
"combinatorial_partial": "./analysis/results/sulfated/phil-bs-native-fit-prior.db",
"combinatorial_grid": "./analysis/results/sulfated/phil-bs-native-grid.db",
"glyspace_unregularized": "./analysis/test/phil-bs-humanish-unregularized.db",
"glyspace_partial": "./analysis/test/phil-bs-humanish-prior.db",
"glyspace_grid": "./analysis/test/phil-bs-humanish-grid.db",
"krambeck_unregularized": "./analysis/test/phil-bs-krambeck-unregularized.db",
"krambeck_partial": "./analysis/test/phil-bs-krambeck-prior.db",
"krambeck_grid": "./analysis/test/phil-bs-krambeck-grid.db"
}
deserializers = {
key: serialize.AnalysisDeserializer(value) for key, value in files.items()
}
print("Loading Datasets")
gcs_map = {
key: chromatogram_tree.ChromatogramFilter(value.query(serialize.GlycanCompositionChromatogram))
for key, value in deserializers.items()
}
all_keys = set()
for group, members in gcs_map.items():
for member in members:
all_keys.add(member.key)
true_positives = set([
gc.glycan_composition for gc in gcs_map["combinatorial_partial"] + gcs_map[
"krambeck_grid"