SeqOptPlot.convergence

SeqOptPlot.convergence(history, figsize=(6, 7))

Plot per-generation convergence: hypervolume, spread and per-objective best.

A multi-panel view of the optimization converging across generations — the dominated hypervolume rising, the front diversity (spread), and each objective’s best front value.

Parameters:

• history (pd.DataFrame) – Per-generation history (SeqOpt.history_ after a run) with a generation column, hypervolume, spread and one best_<objective> column per objective.

• figsize (tuple, default=(6, 7)) – Figure size.

Returns:

• fig (matplotlib.figure.Figure) – The figure.

• ax (numpy.ndarray of matplotlib.axes.Axes) – The panel axes (hypervolume, spread, per-objective best).

Examples

import numpy as np, pandas as pd
import matplotlib.pyplot as plt
from sklearn.ensemble import RandomForestClassifier
import aaanalysis as aa
aa.options["verbose"] = False

# Gamma-secretase (GSEC) substrate data + the bundled interpretable CPP feature set.
df_feat = aa.load_features(name="DOM_GSEC")           # 150 CPP features (with positions, feat_importance)
df_seq  = aa.load_dataset(name="DOM_GSEC", n=50)      # 100 TMD sequences, label 1 = GSEC substrate
labels  = df_seq["label"].to_list()

# A simple RandomForest substrate classifier on the CPP feature matrix.
sf = aa.SequenceFeature()
X = np.asarray(sf.feature_matrix(features=df_feat["feature"],
                                 df_parts=sf.get_df_parts(df_seq=df_seq),
                                 df_scales=aa.load_scales()), dtype=float)
model = RandomForestClassifier(n_estimators=100, random_state=0).fit(X, labels)

# Pick a NON-substrate as the wild-type and design a "super substrate": mutate its TMD to
# maximize the predicted substrate probability with as few mutations as possible.
wt = df_seq[df_seq["label"] == 0].iloc[[0]].reset_index(drop=True)
objectives = [("substrate", "max", "delta_pred"),     # raise P(GSEC substrate) (RF prediction shift)
              ("parsimony", "min", "n_mut")]          # with as few mutations as possible

objectives3 = [("substrate", "max", "delta_pred"),    # raise predicted substrate probability
               ("stability", "min", "delta_cpp"),     # keep the feature profile close to natural
               ("parsimony", "min", "n_mut")]         # few mutations
seqopt = aa.SeqOpt(mode="importance", model=model, target_class=1, random_state=42)
df_pareto = seqopt.run(df_seq=wt, df_feat=df_feat, objectives=objectives3,
                       pop_size=40, n_gen=20, n_mut_max=5, region="tmd")
aa.display_df(df_pareto, n_rows=10, show_shape=True)

DataFrame shape: (28, 9)

	entry	variant	n_mut	sequence_mut	substrate	stability	parsimony	rank	crowding
1	Q14802		0	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	0.000000	0.000000	0.000000	0	inf
2	Q14802	G52L+I55T+S58R+A59R	4	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	36.000000	7.264650	4.000000	0	inf
3	Q14802	V39P+A50K+I55V+S58R+A59R	5	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	30.000000	5.449390	5.000000	0	inf
4	Q14802	S58Q+A59R	2	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	21.000000	3.693250	2.000000	0	0.140833
5	Q14802	S58R	1	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	14.000000	1.549330	1.000000	0	0.138603
6	Q14802	I55V+S58Q+A59R	3	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	25.000000	3.969830	3.000000	0	0.135056
7	Q14802	S58V	1	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	8.000000	0.616080	1.000000	0	0.120829
8	Q14802	A50K+S58Q	2	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	8.000000	0.472060	2.000000	0	0.109567
9	Q14802	G52L+S58R+A59R	3	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	34.000000	6.589060	3.000000	0	0.093956
10	Q14802	G52W+S58R+A59R	3	MQKVTLGLLVFLAGF...PGETPPLITPGSAQS	31.000000	6.094060	3.000000	0	0.087474

aa.plot_settings()
aa.SeqOptPlot().convergence(history=seqopt.history_)
plt.tight_layout(); plt.show()

# figsize sets the multi-panel figure size in inches
aa.plot_settings()
aa.SeqOptPlot().convergence(history=seqopt.history_, figsize=(6, 8))
plt.tight_layout(); plt.show()