API

This Application Programming Interface (API) is the public interface for the objects and functions of our AAanalysis Python toolkit, which can be imported by:

import aaanalysis as aa

You can then access all methods and objects via the aa alias, such as aa.load_dataset.

Data Handling

load_dataset([name, n, random, ...])	Load protein benchmarking datasets.
load_scales([name, just_aaindex, ...])	Load amino acid scales or their classification (AAontology).
load_features([name])	Load feature sets for protein benchmarking datasets.
read_fasta(file_path[, col_id, col_seq, ...])	Read a FASTA file into a DataFrame.
to_fasta([df_seq, file_path, col_id, ...])	Write sequence DataFrame to a FASTA file.
SequencePreprocessor()	Preprocessing class for representing protein sequences as numeric inputs [Breimann25].
StructurePreprocessor([verbose])	Preprocessing class ([pro], requires aaanalysis[pro]) for protein structure features (PDB / CIF / AlphaFold).
EmbeddingPreprocessor([verbose])	Preprocessing class for protein language model (PLM) embeddings.
AnnotationPreprocessor([verbose])	Preprocessing class ([pro], requires aaanalysis[pro]) for per-residue post-translational modification (PTM) / functional-site annotations.
combine_dict_nums([dict_nums])	Concatenate multiple per-residue dict_num inputs along the D axis.

Sequence Analysis

AAlogo([logo_type])	Amino Acid logo (AAlogo) class for computing sequence logo matrices and conservation scores.
AAlogoPlot([logo_type, jmd_n_len, ...])	Amino Acid logo Plot (AAlogoPlot) class for visualizing sequence logos.
AAWindowSampler([verbose, random_state, ...])	Utility class for sampling amino-acid windows / segments from full protein sequences.
comp_seq_sim([seq1, seq2, df_seq])	Compute pairwise similarity between two or more sequences ([pro], requires aaanalysis[pro]).
filter_seq([df_seq, method, ...])	Redundancy reduction of sequences using clustering-based algorithms ([pro], requires aaanalysis[pro]).
scan_motif([df_seq, pos_col, n, ...])	Scan candidate proteins for statistically significant Position Weight Matrix (PWM) occurrences using FIMO ([pro], requires aaanalysis[pro]).

Feature Engineering

AAclust([model_class, model_kwargs, ...])	Amino Acid clustering (AAclust) class: A k-optimized clustering wrapper for selecting redundancy-reduced sets of numerical scales [Breimann24a].
AAclustPlot([model_class, model_kwargs, ...])	Plotting class for AAclust (Amino Acid clustering) results, providing dimensionality-reduction scatter plots, correlation heatmaps, and clustering evaluation charts [Breimann24a].
SequenceFeature([verbose])	Utility feature engineering class using sequences to create CPP feature components (Parts, Splits, and Scales) and data structures [Breimann25].
NumericalFeature()	Utility feature engineering class to process and filter numerical data structures, such as amino acid scales or a feature matrix.
CPP([df_parts, split_kws, df_scales, ...])	Comparative Physicochemical Profiling (CPP) class to create and filter features that are most discriminant between two sets of sequences [Breimann25].
CPPGrid([df_seq, labels, dict_num, ...])	Grid-style sweep over Comparative Physicochemical Profiling (CPP) configurations (Tool) [Breimann25].
CPPPlot([df_scales, df_cat, jmd_n_len, ...])	Plotting class for CPP (Comparative Physicochemical Profiling) results [Breimann25].

PU Learning

dPULearn([model_kwargs, verbose, random_state])	Deterministic Positive-Unlabeled Learning (dPULearn) class for identifying reliable negatives from unlabeled data [Breimann25].
dPULearnPlot()	Plotting class for dPULearn (deterministic Positive-Unlabeled Learning) results [Breimann25].

Explainable AI

TreeModel([list_model_classes, ...])	Tree Model class: A wrapper for tree-based models to obtain Monte Carlo estimates of feature importance and predictions [Breimann25].
ShapModel([explainer_class, ...])	SHAP Model class ([pro], requires aaanalysis[pro]): A wrapper for SHAP (SHapley Additive exPlanations) [Lundberg20] explainers to obtain Monte Carlo estimates for feature impact [Breimann25].

Protein Design

AAMut([verbose, df_scales])	Amino Acid Mutator (AAMut) class for analyzing the physicochemical impact of amino acid substitutions on property scales [Breimann24a].
AAMutPlot([verbose, df_scales])	Plotting class for AAMut (Amino Acid Mutator) results [Breimann24a].
SeqMut([verbose, df_scales])	Sequence Mutator (SeqMut) class for CPP-guided sequence mutation and ΔCPP analysis [Breimann24a].
SeqMutPlot([verbose])	Plotting class for SeqMut (Sequence Mutator) results [Breimann24a].

Utility Functions

comp_auc_adjusted([X, labels, label_test, ...])	Compute an adjusted Area Under the Curve (AUC) [-0.5, 0.5] assessing the similarity between two groups.
comp_bic_score([X, labels])	Compute an adjusted Bayesian Information Criterion (BIC) (-∞, ∞) for assessing clustering quality.
comp_bootstrap_ci([values, n_rounds, ci, seed])	Compute a percentile bootstrap Confidence Interval (CI) of the mean.
comp_detection_metrics([list_scores, ...])	Compute pooled detection metrics at a fixed score threshold.
comp_kld([X, labels, label_test, label_ref])	Compute the Kullback-Leibler Divergence (KLD) [0, ∞) for assessing the similarity between two groups.
comp_per_protein_ap([list_scores, ...])	Compute per-protein average precision (AP) for windowed site prediction.
comp_smooth_scores([scores, method, window, ...])	Smooth a per-residue score vector with a NaN-aware, peak-preserving kernel.
display_df([df, max_width_pct, max_height, ...])	Display DataFrame with specific style as HTML output for jupyter notebooks.
options	A class for managing system-level settings for AAanalysis.
plot_gcfs([option])	Get the current font size (or axes linewidth).
plot_get_cdict([name])	Get color dictionaries specified for AAanalysis.
plot_get_clist([n_colors])	Get a manually curated list of 2 to 9 colors or 'husl' palette for more than 9 colors.
plot_get_cmap([name, n_colors, facecolor_dark])	Get colormaps specified for AAanalysis.
plot_legend([ax, dict_color, list_cat, ...])	Set an independently customizable plot legend.
plot_rank([df_rank, col_score, col_group, ...])	Plot a per-protein rank scatter: max-score-per-protein sorted by score, colored by group.
plot_settings([font_scale, font, ...])	Configure general plot settings.