"distance":{"type":"numerical"},
"affinity": {"colnames": ("ets1_score","runx1_score")}
}, label_map={'cooperative': 1, 'independent': 0})
).run_all(),
"distance,orientation,strength":
BestModel(clf="sklearn.ensemble.RandomForestClassifier",
param_grid=rf_param_grid,
train_data=ct.get_training_df({
"distance":{"type":"numerical"},
"affinity": {"colnames": ("ets1_score","runx1_score")},
"orientation": {"relative":False, "pos_cols": {"ets1_pos":"ets1_ori", "runx1_pos":"runx1_ori"}}
}, label_map={'cooperative': 1, 'independent': 0})
).run_all(),
}
pl.plot_model_metrics(best_models, path="%s/model/pr.png" % basepath, cvfold=10, score_type="pr", varyline=True, title="Average ROC Curves for Ets1-Runx1")
feature_dict = {
"distance":{"type":"numerical"},
"affinity": {"colnames": ("ets1_score","runx1_score")},
"orientation": {"relative":False, "pos_cols": {"ets1_pos":"ets1_ori", "runx1_pos":"runx1_ori"}}
}
train = ct.get_feature_all(feature_dict)
label = ct.get_numeric_label({'cooperative': 1, 'independent': 0})
rf = best_models["distance,orientation,strength"][1]
rf.fit(train,label)
model_name = "%s/model/ets1_runx1_rfmodel.sav" % basepath
pickle.dump(rf, open(model_name, 'wb'))
print("Model saved in %s" % model_name)
# tree.export_graphviz(m.estimators_[5], out_file='tree.dot',
rf_param_grid = {
'n_estimators': [1000], #[500,750,1000],
'max_depth': [10], #[5,10,15],
"min_samples_leaf": [10], #[5,10,15],
"min_samples_split": [10] #[5,10,15]
}
best_models = {
"all":
BestModel(clf="sklearn.ensemble.RandomForestClassifier",
param_grid=rf_param_grid,
train_data=ct.get_training_df(feature_dict)).run_all()
}
pl.plot_model_metrics(best_models,
cvfold=10,
score_type="auc",
varyline=True,
title="Average ROC Curves for Runx1-Ets1")
# # This is usually made based on the best model
rf = ensemble.RandomForestClassifier(n_estimators=1000,
max_depth=10,
min_samples_leaf=10,
min_samples_split=10)
rf.fit(train, label)
model_name = "dist_ori_12merimads.sav"
pickle.dump(rf, open(model_name, 'wb'))
print("Model saved in %s" % model_name)
# tree.export_graphviz(m.estimators_[5], out_file='tree.dot',
# feature_names = xt_df.columns,
"dist,ori,shape_ori":
BestModel(clf="sklearn.ensemble.RandomForestClassifier",
param_grid=rf_param_grid,
train_data=cooptr.get_training_df({
"distance": {
"type": "numerical"
},
"shape_in": {
"seqin": 4,
"smode": "strength",
"direction": "orientation",
"positive_cores": ["GGAA", "GGAT"]
},
"shape_ou": {
"seqin": -4,
"smode": "strength",
"direction": "orientation",
"positive_cores": ["GGAA", "GGAT"]
},
"orientation": {
"positive_cores": ["GGAA", "GGAT"],
"one_hot": True
}
}),
topn=10).run_all()
}
pl.plot_model_metrics(best_models,
cvfold=10,
score_type="auc",
varyline=True)
"one_hot": one_hot_ori,
"pos_cols": {
"%s_pos" % s1: "%s_ori" % s1,
"%s_pos" % s2: "%s_ori" % s2
}
},
},
label_map={
'cooperative': 1,
'independent': 0
})).run_all(),
}
pl.plot_model_metrics(best_models,
path="%s/model/auc_posfeatures.png" % basepath,
cvfold=10,
score_type="auc",
varyline=True,
title="AUC Shape features")
rf = best_models["distance,orientation,shape"][1]
train = ct.get_feature_all({
"distance": {
"type": "numerical"
},
"orientation": {
"relative": rel_ori,
"one_hot": one_hot_ori,
"pos_cols": {
"%s_pos" % s1: "%s_ori" % s1,
"%s_pos" % s2: "%s_ori" % s2
}
"one_hot": True,
"pos_cols": {
"site_str_pos": "site_str_ori",
"site_wk_pos": "site_wk_ori"
}
}
},
label_map={
'cooperative': 1,
'independent': 0
})).run_all()
}
pl.plot_model_metrics(best_models,
path="output/Ets1Ets1/model/auc_all.png",
cvfold=10,
score_type="auc",
varyline=True,
title="Average ROC Curves for Ets1-Ets1")
rf = best_models["distance,orientation,strength"][1]
train = ct.get_feature_all({
"distance": {
"type": "numerical"
},
"affinity": {
"colnames": ("site_str_score", "site_wk_score")
},
"orientation": {
"relative": True,
"one_hot": True,
"colnames": ("ets_score", "runx_score")
},
"orientation": {
"relative": False,
"pos_cols": {
"ets_pos": "ets_ori",
"runx_pos": "runx_ori"
}
}
})).run_all(),
}
pl.plot_model_metrics(
best_models,
cvfold=10,
score_type="auc",
varyline=True,
title=
"Average ROC Curves for Ets1-Runx1\n(using shape and sequence features)"
)
# feature_dict = {
# "distance":{"type":"numerical"},
# "affinity": {"colnames": ("ets_score","runx_score")},
# "orientation": {"relative":False, "pos_cols": {"ets_pos":"ets_ori", "runx_pos":"runx_ori"}}
# }
# train = ct.get_feature_all(feature_dict)
# label = ct.get_numeric_label({'cooperative': 1, 'additive': 0})
# rf = ensemble.RandomForestClassifier(n_estimators=500, max_depth=5, min_samples_leaf=10, min_samples_split=10)
# m = rf.fit(train.values.tolist(),label)
#
# tree.export_graphviz(m.estimators_[5], out_file='tree.dot',
"direction": "inout"
}
}),
).run_all(score_type=score_type),
"dist,shape-top10":
BestModel(
clf="sklearn.ensemble.RandomForestClassifier",
param_grid=rf_param_grid,
train_data=curct.get_training_df({
"distance": {
"type": "numerical"
},
"shape_in": {
"seqin": 4,
"smode": "positional",
"direction": "inout"
}, # maximum seqin is 4
"shape_out": {
"seqin": -4,
"smode": "positional",
"direction": "inout"
}
}),
topn=10).run_all(score_type=score_type),
}
oriname = ori.replace("/", "")
pl.plot_model_metrics(best_models,
cvfold=10,
score_type=score_type,
plotname="auc_%s.png" % oriname)