seismicrna.graph package

Submodules

class seismicrna.graph.abundance.ClusterAbundanceGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: OneTableGraph

Abundance of each cluster.

property col_tracks

property data: Data of the graph.

property details: Additional details about the graph.

get_traces(): Data traces of the graph.

classmethod graph_kind(): Kind of graph.

property path_subject: Subject of the graph.

property predicate: Predicate of the graph.

property row_tracks

classmethod what(): What is being graphed.

property x_title: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.abundance.ClusterAbundanceRunner

Bases: OneTableRunner, AbundanceTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.abundance.ClusterAbundanceWriter(table: Table, **kwargs)

Bases: OneTableWriter

get_graph(**kwargs): Return a graph instance.

iter_graphs(**kwargs): Yield every graph.

class seismicrna.graph.aucroll.RollingAUCGraph(*, fold_table_region: bool, struct_file: Path | None, struct_reg: str | None, terminal_pairs: bool, branch: str, **kwargs)

Bases: StructOneTableGraph, RollingGraph

property data: Data of the graph.

get_traces(): Data traces of the graph.

classmethod graph_kind(): Kind of graph.

property profile_names: Names of the profiles as they appear in the data.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.aucroll.RollingAUCRunner

Bases: RollingRunner, StructOneTableRunner, PositionTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.aucroll.RollingAUCWriter(table: Table, **kwargs)

Bases: StructOneTableWriter, TableWriter

get_graph(rels_group: str, **kwargs): Return a graph instance.

class seismicrna.graph.base.Annotation(row: int, col: int, x: float, y: float, text: str, **kwargs)

Bases: object

Text annotation in a graph.

class seismicrna.graph.base.BaseGraph

Bases: ABC

Base class for all types of graphs.

property annotations: list[Annotation]: Text annotations for the figure.

abstract property branches: dict[str, str]: Branches of the workflow.

property data: pd.DataFrame: Data of the graph.

property details: list[str]: Additional details about the graph.

property figure: Figure object.

get_path(ext: str): Path to the output file of the graph.

get_path_fields(): Path fields.

classmethod get_path_segs(): Path segments.

abstractmethod get_traces() → Iterable[tuple[tuple[int, int], go.Trace]]: Data traces of the graph.

property graph_filename: Name of the graph’s output file, without its extension.

abstractmethod classmethod graph_kind() → str: Kind of graph.

abstract property path_subject: str: Subject of the graph.

property predicate: list[str]: Predicate of the graph.

abstract property ref: str: Name of the reference sequence from which the data come.

abstract property reg: str: Name of the reference region from which the data come.

abstract property sample: str: Name(s) of the sample(s) from which the data come.

abstract property seq: DNA: Sequence of the region from which the data come.

property title: Title of the graph.

abstract property title_action_sample: str: Action and sample for the title.

abstract property top: Path: Path of the top-level output directory for all files.

abstractmethod classmethod what() → str: What is being graphed.

write(csv: bool, html: bool, force: bool = False): Write the filtered files.

write_csv(force: bool): Write the graph’s source data to a CSV file.

write_html(force: bool): Write the graph to an HTML file.

abstract property x_title: str: Title of the x-axis.

abstract property y_title: str: Title of the y-axis.

class seismicrna.graph.base.BaseRunner

Bases: ABC

classmethod get_cmd() → str: Name of the command that graphs this kind of graph, which is the name of the module that defines the runner, as in every graph module’s own COMMAND.

abstractmethod classmethod get_input_loader() → Callable[[...], Generator]: Function to load input files.

classmethod get_var_params() → list[Argument | Option]: Parameters that can vary among different classes.

abstractmethod classmethod get_writer_type() → type[BaseWriter]: Type of Writer.

classmethod load_input_files(input_path: Iterable[str | Path], **kwargs): Find, filter, and load all input files.

classmethod params() → list[Argument | Option]: Parameters for the command line.

abstractmethod classmethod run(*args, **kwargs) → list[Path]: Run graphing.

classmethod universal_input_params(): Universal parameters controlling the input data.

classmethod universal_output_params(): Universal parameters controlling the output graph.

class seismicrna.graph.base.BaseWriter

Bases: ABC

Write the proper type(s) of graph.

abstractmethod iter_graphs(*args, **kwargs) → Generator[BaseGraph, None, None]: Yield every graph.

write(*args, csv: bool, html: bool, force: bool, **kwargs): Generate and write every type of graph.

seismicrna.graph.base.get_action_name(source: MutsDataset | Table)

seismicrna.graph.base.make_path_subject(action: str, k: int | None, clust: int | None)

Build the path subject string that identifies a graph’s data source.

Parameters:

action (str) – Action name (one of ACTION_IDMUT, ACTION_FILTER, or ACTION_CLUSTER).
k (int or None) – Number of clusters; must be None or 0 for non-cluster actions.
clust (int or None) – Cluster index; must be None or 0 for non-cluster actions.

Returns:

A string suitable for use as the subject component of a file path (e.g. "all", "filtered", or "clustered-2-1").

Return type:

str

seismicrna.graph.base.make_title_action_sample(action: str, sample: str)

class seismicrna.graph.cgroup.ClusterGroupGraph

Bases: BaseGraph, ABC

Graph in which clusters can be placed in subplots.

property clust_proportions: Mapping from each (k, cluster) to its proportion, or None if cluster proportions are not available for this graph.

property col_proportions: Cluster proportions for the column tracks.

property col_titles: Titles of the columns.

abstract property col_tracks: list[tuple[int, int]] | None: Track for each column of subplots.

property ncols: Number of columns of subplots.

property nrows: Number of rows of subplots.

property row_proportions: Cluster proportions for the row tracks.

property row_titles: Titles of the rows.

abstract property row_tracks: list[tuple[int, int]] | None: Track for each row of subplots.

class seismicrna.graph.cgroup.ClusterGroupRunner

Bases: BaseRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

seismicrna.graph.cgroup.cgroup_table(source: Dataset | Table, cgroup: str)

seismicrna.graph.cgroup.get_clust_proportions(top, sample, branches, ref, reg): Load the proportion of each cluster, keyed by (k, cluster), from the cluster abundance table identified by these path fields. Returns None if no such abundance table exists (e.g. the source is not clustered).

seismicrna.graph.cgroup.get_ks(source: Dataset | Table): List the numbers of clusters for a source of data.

seismicrna.graph.cgroup.get_ks_clusts(source: Dataset | Table): List the clusters for a source of data.

seismicrna.graph.cgroup.make_tracks(source: Dataset | Table, k: int | None, clust: int | None, *, k_clust_list: list[tuple[int, int]] | None = None): Make an index for the rows or columns of a graph.

class seismicrna.graph.color.ColorMap(name: str, **kwargs)

Bases: ABC

Color map for a graph.

get(item: Hashable, default: Any | None = None)

abstractmethod get_default_color() → str: Default color.

class seismicrna.graph.color.ColorMapGraph(*, cmap: str | None = None, **kwargs)

Bases: BaseGraph, ABC

Graph with an explicit color map.

property cmap: ColorMap: Color map of the graph.

abstractmethod classmethod get_cmap_type() → type[ColorMap]: Type of the color map.

class seismicrna.graph.color.RelColorMap(name: str, v: str, n: str, e: str, m: str, d: str, i: str, s: str, a: str, c: str, g: str, t: str, x: str)

Bases: ColorMap

Color map for relationships.

get_default_color(): Default color.

class seismicrna.graph.color.SeqColorMap(name: str, a: str, c: str, g: str, t: str, n: str)

Bases: ColorMap

Color map for bases A, C, G, and T.

get_default_color(): Default color.

seismicrna.graph.color.get_cmap(cmap_class: type[ColorMap], name: str | None = None): Get a color map of a given class by its name.

seismicrna.graph.color.get_colormaps(cmap_class: type[ColorMap]): Return a dict of all color maps of a given class.

class seismicrna.graph.corroll.RollingCorrelationGraph(*, metric: str, **kwargs)

Bases: TwoTableMergedClusterGroupGraph, OneRelGraph, RollingGraph

property details: Additional details about the graph.

classmethod graph_kind(): Kind of graph.

property predicate: Predicate of the graph.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.corroll.RollingCorrelationRunner

Bases: TwoTableRelClusterGroupRunner, RollingRunner, PositionTableRunner

classmethod get_var_params(): Parameters that can vary among different classes.

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write comparison graphs for each pair.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
compself (bool) – If True, compare each table with itself.
comppair (bool) – If True, compare every pair of different tables with the same reference and region.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.corroll.RollingCorrelationWriter(table1: Table, table2: Table, **kwargs)

Bases: TwoTableRelClusterGroupWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.dataset.DatasetGraph(*, dataset: MutsDataset, num_cpus: int, **kwargs)

Bases: OneRelGraph, OneSourceClusterGroupGraph, ABC

Graph based on one Dataset.

property action: Action that generated the data.

property branches: Branches of the workflow.

property clust_proportions: Proportion of each cluster, keyed by (k, cluster), loaded from the cluster abundance table that accompanies this dataset; None if the dataset is not clustered or no abundance table exists.

property details: Additional details about the graph.

property path_subject: Subject of the graph.

property pattern: Relationship pattern for the graph.

property predicate: Predicate of the graph.

property ref: Name of the reference sequence from which the data come.

property reg: Name of the reference region from which the data come.

property sample: Name(s) of the sample(s) from which the data come.

property seq: Sequence of the region from which the data come.

property title_action_sample: Action and sample for the title.

property top: Path of the top-level output directory for all files.

class seismicrna.graph.dataset.DatasetRunner

Bases: RelRunner, ClusterGroupRunner, ABC

classmethod get_input_loader(): Function to load input files.

classmethod get_var_params(): Parameters that can vary among different classes.

abstractmethod classmethod get_writer_type() → type[DatasetWriter]: Type of Writer.

classmethod run(input_path: Iterable[str | Path], *, verify_times: bool, num_cpus: int, **kwargs)

Load all datasets and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for datasets.
verify_times (bool) – Whether to verify file modification times when loading datasets.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.dataset.DatasetWriter(*, dataset: MutsDataset, **kwargs)

Bases: BaseWriter, ABC

abstractmethod get_graph(*args, **kwargs) → DatasetGraph: Return a graph instance.

iter_graphs(*, rels: list[str], cgroup: str, **kwargs)

Yield graph instances for every relationship and cluster group.

Parameters:

rels (list[str]) – Relationship codes to graph.
cgroup (str) – Cluster-grouping strategy.
**kwargs – Additional keyword arguments forwarded to get_graph.

Yields:

DatasetGraph – One graph per (cluster group, relationship) combination.

class seismicrna.graph.delprof.DeltaProfileGraph(*, k1: int | None, clust1: int | None, k2: int | None, clust2: int | None, **kwargs)

Bases: TwoTableMergedClusterGroupGraph, OneRelGraph, ColorMapGraph

classmethod get_cmap_type(): Type of the color map.

classmethod graph_kind(): Kind of graph.

classmethod what(): What is being graphed.

property x_title: str: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.delprof.DeltaProfileRunner

Bases: TwoTableRelClusterGroupRunner, PositionTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write comparison graphs for each pair.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
compself (bool) – If True, compare each table with itself.
comppair (bool) – If True, compare every pair of different tables with the same reference and region.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.delprof.DeltaProfileWriter(table1: Table, table2: Table, **kwargs)

Bases: TwoTableRelClusterGroupWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.giniroll.RollingGiniGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: RollingStatGraph

classmethod graph_kind(): Kind of graph.

classmethod stat_func(): Function to compute a statistic on the data.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.giniroll.RollingGiniRunner

Bases: RollingStatRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.giniroll.RollingGiniWriter(table: Table, **kwargs)

Bases: RollingStatWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.hist.HistogramGraph(*, hist_bins: int, hist_margin: float, **kwargs)

Bases: OneTableRelClusterGroupGraph, MultiRelsGraph, ColorMapGraph, ABC

Histogram of relationship(s) in one table.

property data: Data of the graph.

property data_header: Header of the filtered data (not of the entire table).

get_bounds(data: pd.DataFrame): Get the lower and upper bounds of the histogram.

classmethod get_cmap_type(): Type of the color map.

get_edges(data: pd.DataFrame): Get the edges of the histogram bins.

get_traces(): Data traces of the graph.

property x_title: Title of the x-axis.

class seismicrna.graph.hist.HistogramRunner

Bases: OneTableRelClusterGroupRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

class seismicrna.graph.hist.HistogramWriter(table: Table, **kwargs)

Bases: OneTableRelClusterGroupWriter, ABC

get_graph(rels_group: str, **kwargs)

Instantiate a HistogramGraph for the given relationship group.

Parameters:

rels_group (str) – Relationship code string for the graph.
**kwargs – Additional keyword arguments forwarded to the graph class.

Returns:

A new graph instance for the given relationship group.

Return type:

HistogramGraph

abstractmethod classmethod get_graph_type() → type[HistogramGraph]: Type of graph.

seismicrna.graph.hist.get_edges_index(edges: np.ndarray, use_ratio: bool)

Generate an index for the edges of histogram bins.

Parameters:

edges (numpy.ndarray) – Edges of histogram bins.
use_ratio (bool) – Assume the edges represent ratios rather than counts.

Returns:

Index for the edges.

Return type:

pandas.Index | pandas.MultiIndex

class seismicrna.graph.histpos.PositionHistogramGraph(*, hist_bins: int, hist_margin: float, **kwargs)

Bases: HistogramGraph

classmethod graph_kind(): Kind of graph.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.histpos.PositionHistogramRunner

Bases: HistogramRunner, PositionTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.histpos.PositionHistogramWriter(table: Table, **kwargs)

Bases: HistogramWriter, TableWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.histread.ReadHistogramGraph(*, hist_bins: int, hist_margin: float, **kwargs)

Bases: HistogramGraph

classmethod graph_kind(): Kind of graph.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.histread.ReadHistogramRunner

Bases: HistogramRunner, ReadTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.histread.ReadHistogramWriter(table: Table, **kwargs)

Bases: HistogramWriter, TableWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.mutdist.MutationDistanceGraph(*, mutdist_null: bool, **kwargs)

Bases: DatasetGraph, ColorMapGraph

Distance between the closest two mutations in each read.

property data: Data of the graph.

property g_test: G-test statistic and P-value.

classmethod get_cmap_type(): Type of the color map.

get_traces(): Data traces of the graph.

classmethod graph_kind(): Kind of graph.

property hists

property loc_clusters

property max_read_length

property row_tracks: Track for each row of subplots.

property table

property tabulator

classmethod what(): What is being graphed.

property x_title: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.mutdist.MutationDistanceRunner

Bases: DatasetRunner

classmethod get_var_params(): Parameters that can vary among different classes.

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all datasets and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for datasets.
verify_times (bool) – Whether to verify file modification times when loading datasets.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.mutdist.MutationDistanceWriter(*, dataset: MutsDataset, **kwargs)

Bases: DatasetWriter

get_graph(rel, **kwargs): Return a graph instance.

seismicrna.graph.mutdist.get_null_name(name: str)

class seismicrna.graph.onesource.OneSourceClusterGroupGraph(*, k: int | None, clust: int | None, k_clust_list: list[tuple[int, int]] | None = None, **kwargs)

Bases: OneSourceGraph, ClusterGroupGraph, ABC

property path_subject: Subject of the graph.

class seismicrna.graph.onesource.OneSourceGraph

Bases: BaseGraph, ABC

Graph of data from one source of data (Dataset or Table).

property action: str: Action that generated the data.

property col_tracks

property title_action_sample: Action and sample for the title.

class seismicrna.graph.onestruct.StructOneTableGraph(*, fold_table_region: bool, struct_file: Path | None, struct_reg: str | None, terminal_pairs: bool, branch: str, **kwargs)

Bases: OneTableRelClusterGroupGraph, OneRelGraph, ABC

Graph of data from one Table applied to RNA structure(s).

property branches: Branches of the workflow.

property details: Additional details about the graph.

get_path_fields(): Path fields.

iter_profiles(): Yield each RNAProfile from the table.

iter_states(): Yield each RNAState.

property path_subject: Subject of the graph.

property predicate: Predicate of the graph.

property struct_reg: Name of the region from which the structure comes.

class seismicrna.graph.onestruct.StructOneTableRunner

Bases: OneTableRelClusterGroupRunner, ABC

classmethod get_input_loader(): Function to load input files.

classmethod get_var_params(): Parameters that can vary among different classes.

class seismicrna.graph.onestruct.StructOneTableWriter(table: Table, **kwargs)

Bases: OneTableRelClusterGroupWriter, ABC

iter_graphs(*, rels: list[str], cgroup: str, struct_file: Iterable[str | Path] = (), branch: str = '', fold_coords: Iterable[tuple[str, int, int]] = (), fold_primers: Iterable[tuple[str, DNA, DNA]] = (), fold_regions_file: str | None = None, fold_table_region: bool = False, **kwargs)

Yield graphs for every relationship, cluster, and structure.

Parameters:

rels (list[str]) – Relationship codes to graph.
cgroup (str) – Cluster-grouping strategy.
struct_file (Iterable[str or Path], optional) – CT files of RNA structures to use directly.
branch (str, optional) – Branch label for the fold step in the output path.
fold_coords (Iterable[tuple[str, int, int]], optional) – (ref, end5, end3) tuples defining structure regions.
fold_primers (Iterable[tuple[str, DNA, DNA]], optional) – (ref, fwd, rev) primer tuples defining structure regions.
fold_regions_file (str or None, optional) – Path to a file of fold region definitions.
fold_table_region (bool, optional) – Whether to use the table’s region as the fold region.
**kwargs – Forwarded to get_graph.

Yields:

StructOneTableGraph – One graph per (cluster group, relationship, structure) combination.

class seismicrna.graph.onetable.OneTableGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: TableGraph, OneSourceGraph, ABC

Graph of data from one Table.

property action: Action that generated the data.

property branches: Branches of the workflow.

property ref: Name of the reference sequence from which the data come.

property reg: Name of the reference region from which the data come.

property sample: Name(s) of the sample(s) from which the data come.

property seq: Sequence of the region from which the data come.

property top: Path of the top-level output directory for all files.

class seismicrna.graph.onetable.OneTableRelClusterGroupGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: OneTableGraph, RelTableGraph, OneSourceClusterGroupGraph, ABC

property clust_proportions: Proportion of each cluster, keyed by (k, cluster), loaded from the cluster abundance table that accompanies this table. Returns None if the table is not clustered or no abundance table exists.

property row_tracks: Track for each row of subplots.

class seismicrna.graph.onetable.OneTableRelClusterGroupRunner: Bases: OneTableRunner, RelTableRunner, ClusterGroupRunner, ABC

class seismicrna.graph.onetable.OneTableRelClusterGroupWriter(table: Table, **kwargs)

Bases: OneTableWriter, ABC

iter_graphs(*, rels: list[str], cgroup: str, **kwargs)

Yield graph instances for every relationship and cluster group.

Parameters:

rels (list[str]) – Relationship code strings to graph.
cgroup (str) – Cluster-grouping strategy.
**kwargs – Additional keyword arguments forwarded to get_graph.

Yields:

OneTableRelClusterGroupGraph – One graph per (cluster group, relationship) combination.

class seismicrna.graph.onetable.OneTableRunner

Bases: TableRunner, ABC

classmethod run(input_path: Iterable[str | Path], *, verify_times: bool, num_cpus: int, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.onetable.OneTableWriter(table: Table, **kwargs)

Bases: TableWriter, ABC

abstractmethod get_graph(*args, **kwargs) → OneTableGraph: Return a graph instance.

property table: The table providing the data for the graph(s).

class seismicrna.graph.poscorr.PositionCorrelationGraph(*, dataset: MutsDataset, num_cpus: int, **kwargs)

Bases: PositionPairGraph, ABC

Phi correlations between pairs of positions.

classmethod get_pair_func(): Function to compare each pair of positions.

classmethod graph_kind(): Kind of graph.

classmethod what(): What is being graphed.

class seismicrna.graph.poscorr.PositionCorrelationRunner

Bases: PositionPairRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all datasets and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for datasets.
verify_times (bool) – Whether to verify file modification times when loading datasets.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.poscorr.PositionCorrelationWriter(*, dataset: MutsDataset, **kwargs)

Bases: PositionPairWriter

classmethod graph_type(): Type of graph.

class seismicrna.graph.pospair.PositionPairGraph(*, dataset: MutsDataset, num_cpus: int, **kwargs)

Bases: DatasetGraph, ABC

Function of pairs of positions.

property data: Data of the graph.

abstractmethod classmethod get_pair_func(): Function to compare each pair of positions.

get_traces(): Data traces of the graph.

property row_tracks: Track for each row of subplots.

property x_title: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.pospair.PositionPairRunner

Bases: DatasetRunner, ABC

abstractmethod classmethod get_writer_type() → type[PositionPairWriter]: Type of Writer.

class seismicrna.graph.pospair.PositionPairWriter(*, dataset: MutsDataset, **kwargs)

Bases: DatasetWriter

get_graph(rel, **kwargs)

Instantiate a PositionPairGraph for the given relationship.

Parameters:

rel (str) – One-letter relationship code.
**kwargs – Additional keyword arguments forwarded to the graph class.

Returns:

A new graph instance for the given relationship.

Return type:

PositionPairGraph

abstractmethod classmethod graph_type() → type[PositionPairGraph]: Type of graph.

class seismicrna.graph.profile.MultiRelsProfileGraph(*, rels: str, **kwargs)

Bases: MultiRelsGraph, ProfileGraph

Stacked bar graph with multiple relationships per position.

classmethod get_cmap_type(): Type of the color map.

get_traces(): Data traces of the graph.

classmethod what(): What is being graphed.

class seismicrna.graph.profile.OneRelProfileGraph(*, rel: str, **kwargs)

Bases: OneRelGraph, ProfileGraph

Bar graph with one relationship per position.

classmethod get_cmap_type(): Type of the color map.

get_traces(): Data traces of the graph.

classmethod what(): What is being graphed.

class seismicrna.graph.profile.ProfileGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: OneTableRelClusterGroupGraph, ColorMapGraph, ABC

Bar graph of a mutational profile for one table.

property data: Data of the graph.

property data_header: Header of the filtered data (not of the entire table).

classmethod graph_kind(): Kind of graph.

property x_title: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.profile.ProfileRunner

Bases: OneTableRelClusterGroupRunner, PositionTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.profile.ProfileWriter(table: Table, **kwargs)

Bases: OneTableRelClusterGroupWriter

get_graph(rels_group: str, **kwargs): Return a graph instance.

class seismicrna.graph.rel.MultiRelsGraph(*, rels: str, **kwargs)

Bases: RelGraph, ABC

Graph of one or more relationships.

property codestring: String of the relationship code(s).

property rel_names: Names of the relationships to graph.

class seismicrna.graph.rel.OneRelGraph(*, rel: str, **kwargs)

Bases: RelGraph, ABC

Graph of exactly one type of relationship.

property codestring: String of the relationship code(s).

property rel_name: Name of the relationship to graph.

property rel_names: Names of the relationships to graph.

class seismicrna.graph.rel.RelGraph

Bases: BaseGraph, ABC

Graph of one or more types of relationships.

abstract property codestring: str: String of the relationship code(s).

abstract property rel_names: Names of the relationships to graph.

property relationships: str: Relationships being graphed as a slash-separated string.

class seismicrna.graph.rel.RelRunner

Bases: BaseRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

class seismicrna.graph.roc.ROCGraph(*, fold_table_region: bool, struct_file: Path | None, struct_reg: str | None, terminal_pairs: bool, branch: str, **kwargs)

Bases: StructOneTableGraph

Graph of a receiver operating characteristic (ROC) curve.

property data: Data of the graph.

property fpr: False positive rate (FPR) of each RNA state.

get_traces(): Data traces of the graph.

classmethod graph_kind(): Kind of graph.

property profile_names: Names of the profiles as they appear in the data.

property tpr: True positive rate (TPR) of each RNA state.

classmethod what(): What is being graphed.

property x_title: Title of the x-axis.

property y_title: Title of the y-axis.

class seismicrna.graph.roc.ROCRunner

Bases: StructOneTableRunner, PositionTableRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.roc.ROCWriter(table: Table, **kwargs)

Bases: StructOneTableWriter, TableWriter

get_graph(rels_group: str, **kwargs): Return a graph instance.

seismicrna.graph.roc.rename_columns(df: pd.DataFrame): Rename the levels of the columns.

class seismicrna.graph.roll.RollingGraph(*, window: int, winmin: int, **kwargs)

Bases: TableGraph, ABC

property details: Additional details about the graph.

property predicate: Predicate of the graph.

property x_title: Title of the x-axis.

class seismicrna.graph.roll.RollingRunner

Bases: TableRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

class seismicrna.graph.scatter.ScatterGraph(*, metric: str, **kwargs)

Bases: TwoTableRelClusterGroupGraph, OneRelGraph, ColorMapGraph

property annotations: Text annotations for the figure.

property data: Data of the graph.

classmethod get_cmap_type(): Type of the color map.

get_traces(): Data traces of the graph.

classmethod graph_kind(): Kind of graph.

classmethod what(): What is being graphed.

property x_title: Title of the x-axis.

property xmax

property y_title: Title of the y-axis.

property ymax

class seismicrna.graph.scatter.ScatterRunner

Bases: TwoTableRelClusterGroupRunner, PositionTableRunner

classmethod get_var_params(): Parameters that can vary among different classes.

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write comparison graphs for each pair.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
compself (bool) – If True, compare each table with itself.
comppair (bool) – If True, compare every pair of different tables with the same reference and region.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.scatter.ScatterWriter(table1: Table, table2: Table, **kwargs)

Bases: TwoTableRelClusterGroupWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.snrroll.RollingSNRGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: RollingStatGraph

classmethod graph_kind(): Kind of graph.

classmethod stat_func(): Function to compute a statistic on the data.

classmethod what(): What is being graphed.

property y_title: Title of the y-axis.

class seismicrna.graph.snrroll.RollingSNRRunner

Bases: RollingStatRunner

classmethod get_writer_type(): Type of Writer.

classmethod run(*args, **kwargs)

Load all tables and write graphs for each.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.snrroll.RollingSNRWriter(table: Table, **kwargs)

Bases: RollingStatWriter

classmethod get_graph_type(): Type of graph.

class seismicrna.graph.statroll.RollingStatGraph(*, table: Table | PositionTable | AbundanceTable, **kwargs)

Bases: OneTableRelClusterGroupGraph, OneRelGraph, RollingGraph, ABC

property data: Data of the graph.

get_traces(): Data traces of the graph.

abstractmethod classmethod stat_func() → Callable[[pd.Series], pd.Series]: Function to compute a statistic on the data.

class seismicrna.graph.statroll.RollingStatRunner: Bases: OneTableRelClusterGroupRunner, RollingRunner, PositionTableRunner, ABC

class seismicrna.graph.statroll.RollingStatWriter(table: Table, **kwargs)

Bases: OneTableRelClusterGroupWriter, TableWriter, ABC

get_graph(rels_group: str, **kwargs)

Instantiate a RollingStatGraph for the given relationship.

Parameters:

rels_group (str) – One-letter relationship code.
**kwargs – Additional keyword arguments forwarded to the graph class.

Returns:

A new rolling-statistic graph for the given relationship.

Return type:

RollingGraph

abstractmethod classmethod get_graph_type() → type[RollingGraph]: Type of graph.

class seismicrna.graph.table.AbundanceTableRunner

Bases: TableRunner, ABC

classmethod get_input_loader(): Function to load input files.

class seismicrna.graph.table.PositionTableRunner

Bases: RelTableRunner, ABC

classmethod get_input_loader(): Function to load input files.

class seismicrna.graph.table.ReadTableRunner

Bases: RelTableRunner, ABC

classmethod get_input_loader(): Function to load input files.

class seismicrna.graph.table.RelTableGraph(*, graph_quantile: float, **kwargs)

Bases: TableGraph, RelGraph, ABC

property details: Additional details about the graph.

property predicate: Predicate of the graph.

class seismicrna.graph.table.RelTableRunner

Bases: RelRunner, TableRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

class seismicrna.graph.table.TableGraph(*, use_ratio: bool, **kwargs)

Bases: BaseGraph, ABC

Graph based on one or more tables.

property data_kind

either “ratio” or “count”.

Type:: Kind of data being used

class seismicrna.graph.table.TableRunner

Bases: BaseRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

abstractmethod classmethod get_writer_type() → type[TableWriter]: Type of Writer.

class seismicrna.graph.table.TableWriter(*tables: Table, **kwargs)

Bases: BaseWriter, ABC

abstractmethod iter_graphs(*args, **kwargs) → Generator[TableGraph, None, None]: Yield every graph.

seismicrna.graph.table.load_abundance_tables(input_paths: Iterable[str | Path], **kwargs): Load read tables.

seismicrna.graph.table.load_pos_tables(input_paths: Iterable[str | Path], **kwargs): Load position tables.

seismicrna.graph.table.load_read_tables(input_paths: Iterable[str | Path], **kwargs): Load read tables.

seismicrna.graph.trace.get_hist_trace(data: pd.Series, rel: str, cmap: ColorMap)

Build a histogram bar trace for one relationship type.

Parameters:

data (pd.Series) – Histogram data. If the index is a pd.MultiIndex its levels are HIST_LOWER_NAME and HIST_UPPER_NAME (ratio bins); otherwise it is a plain pd.Index named HIST_COUNT_NAME (count bins).
rel (str) – Name of the relationship (used as the trace name and for color lookup).
cmap (ColorMap) – Color map used to look up the color for rel.

Returns:

A bar trace representing the histogram for rel.

Return type:

plotly.graph_objects.Bar

seismicrna.graph.trace.get_line_trace(data: pd.Series, cluster: str)

seismicrna.graph.trace.get_pairwise_position_trace(data: pd.Series, end5: int, end3: int)

Build a pairwise-position heatmap trace.

Parameters:

data (pd.Series) – Long-form data with a two-level pd.MultiIndex of (position_a, position_b) pairs and values representing the pairwise statistic.
end5 (int) – 5’ end position of the region (inclusive, 1-based).
end3 (int) – 3’ end position of the region (inclusive, 1-based).

Returns:

A symmetric heatmap trace covering positions end5 to end3.

Return type:

plotly.graph_objects.Heatmap

seismicrna.graph.trace.get_roc_trace(fpr: pd.Series, tpr: pd.Series, profile: str, struct: str)

Build an ROC curve trace for one profile/structure combination.

Parameters:

fpr (pd.Series) – False positive rates.
tpr (pd.Series) – True positive rates.
profile (str) – Name of the mutational profile.
struct (str) – Name or identifier of the RNA structure.

Returns:

An ROC curve trace with AUC annotated in the trace name.

Return type:

plotly.graph_objects.Scatter

seismicrna.graph.trace.get_rolling_auc_trace(auc: pd.Series, profile: str, struct: str)

seismicrna.graph.trace.get_seq_base_bar_trace(data: pd.Series, cmap: ColorMap, base: str)

Build a bar trace for one DNA base type.

Parameters:

data (pd.Series) – Data indexed by a MultiIndex with levels including BASE_NAME and POS_NAME.
cmap (ColorMap) – Color map used to look up the color for base.
base (str) – Single-character DNA base code (A, C, G, T, or N).

Returns:

A bar trace for the positions of the given base type.

Return type:

plotly.graph_objects.Bar

seismicrna.graph.trace.get_seq_base_scatter_trace(xdata: pd.Series, ydata: pd.Series, cmap: ColorMap, base: str)

Build a scatter trace for one DNA base type.

Parameters:

xdata (pd.Series) – x-axis data indexed by a MultiIndex with levels including BASE_NAME and POS_NAME.
ydata (pd.Series) – y-axis data with the same index as xdata.
cmap (ColorMap) – Color map used to look up the color for base.
base (str) – Single-character DNA base code (A, C, G, T, or N).

Returns:

A scatter trace for the positions of the given base type.

Return type:

plotly.graph_objects.Scatter

seismicrna.graph.trace.get_seq_line_trace(data: pd.Series)

seismicrna.graph.trace.get_seq_stack_bar_trace(data: pd.Series, rel: str, cmap: ColorMap)

Build a stacked bar trace for one relationship type.

Parameters:

data (pd.Series) – Per-position data indexed by a MultiIndex with levels including BASE_NAME and POS_NAME.
rel (str) – Name of the relationship (used as the trace name and for color lookup).
cmap (ColorMap) – Color map used to look up the color for rel.

Returns:

A bar trace for all positions, colored by relationship type.

Return type:

plotly.graph_objects.Bar

seismicrna.graph.trace.iter_hist_traces(data: pd.DataFrame, cmap: ColorMap)

seismicrna.graph.trace.iter_line_traces(lines: pd.DataFrame)

seismicrna.graph.trace.iter_roc_traces(fprs: pd.DataFrame, tprs: pd.DataFrame, profile: str)

Yield one ROC trace per RNA structure.

Parameters:

fprs (pd.DataFrame) – DataFrame whose columns correspond to structures and whose values are false positive rates.
tprs (pd.DataFrame) – DataFrame whose columns correspond to structures and whose values are true positive rates; columns must match fprs.
profile (str) – Name of the mutational profile.

Yields:

plotly.graph_objects.Scatter – One ROC trace per structure column.

seismicrna.graph.trace.iter_rolling_auc_traces(aucs: pd.DataFrame, profile: str)

seismicrna.graph.trace.iter_seq_base_bar_traces(data: pd.Series, cmap: ColorMap)

seismicrna.graph.trace.iter_seq_base_scatter_traces(xdata: pd.Series, ydata: pd.Series, cmap: ColorMap)

Yield one scatter trace per DNA base type.

Parameters:

xdata (pd.Series) – x-axis data indexed by a MultiIndex including BASE_NAME and POS_NAME.
ydata (pd.Series) – y-axis data with the same index as xdata.
cmap (ColorMap) – Color map for base coloring.

Yields:

plotly.graph_objects.Scatter – One trace per base in DNA.alph().

seismicrna.graph.trace.iter_seq_line_traces(data: pd.Series, *_, **__)

Yield a single line trace for sequence data.

Parameters:

data (pd.Series) – Per-position data indexed by a MultiIndex including POS_NAME.
*_ – Ignored extra arguments (for call-signature compatibility).
**__ – Ignored extra arguments (for call-signature compatibility).

Yields:

plotly.graph_objects.Scatter – One line trace for the entire series.

seismicrna.graph.trace.iter_seqbar_stack_traces(data: pd.DataFrame, cmap: ColorMap)

seismicrna.graph.trace.iter_stack_bar_traces(data: pd.DataFrame)

class seismicrna.graph.twotable.TwoTableGraph(*, out_dir: str | Path, table1: Table | PositionTable, table2: Table | PositionTable, **kwargs)

Bases: TableGraph, ABC

Graph of two Tables.

property action1: Action that generated dataset 1.

property action2: Action that generated dataset 2.

property action_sample1: Action and sample of dataset 1.

property action_sample2: Action and sample of dataset 2.

property branches: Branches of the workflow.

property ref: Name of the reference sequence from which the data come.

property reg: Name of the reference region from which the data come.

property sample: Name(s) of the sample(s) from which the data come.

property sample1: Name of sample 1.

property sample2: Name of sample 2.

property seq: Sequence of the region from which the data come.

property title_action_sample: Action and sample for the title.

property top: Path of the top-level output directory for all files.

class seismicrna.graph.twotable.TwoTableMergedClusterGroupGraph(*, k1: int | None, clust1: int | None, k2: int | None, clust2: int | None, **kwargs)

Bases: TwoTableRelClusterGroupGraph, ABC

Graph of a pair of datasets over the same sequence in which the data series are merged in some fashion into another series, and the original data are not graphed directly.

property data: Data of the graph.

get_traces(): Data traces of the graph.

class seismicrna.graph.twotable.TwoTableRelClusterGroupGraph(*, k1: int | None, clust1: int | None, k2: int | None, clust2: int | None, **kwargs)

Bases: TwoTableGraph, RelTableGraph, ClusterGroupGraph, ABC

property col_proportions: Cluster proportions for the column tracks.

property col_tracks: Track for each column of subplots.

property data1: Data from table 1.

property data2: Data from table 2.

property path_subject: Subject of the graph.

property path_subject1: Name of subject 1.

property path_subject2: Name of subject 2.

property row_proportions: Cluster proportions for the row tracks.

property row_tracks: Track for each row of subplots.

class seismicrna.graph.twotable.TwoTableRelClusterGroupRunner: Bases: TwoTableRunner, RelTableRunner, ClusterGroupRunner, ABC

class seismicrna.graph.twotable.TwoTableRelClusterGroupWriter(table1: Table, table2: Table, **kwargs)

Bases: TwoTableWriter, ABC

iter_graphs(*, rels: list[str], cgroup: str, **kwargs)

Yield graphs for every relationship and cluster-group pair.

Parameters:

rels (list[str]) – Relationship codes to graph.
cgroup (str) – Cluster-grouping strategy applied to both tables.
**kwargs – Additional keyword arguments forwarded to the graph class.

Yields:

TwoTableRelClusterGroupGraph – One graph per (cluster group from table1, cluster group from table2, relationship) combination.

class seismicrna.graph.twotable.TwoTableRunner

Bases: TableRunner, ABC

classmethod get_var_params(): Parameters that can vary among different classes.

abstractmethod classmethod get_writer_type() → type[TwoTableWriter]: Type of Writer.

classmethod run(input_path: Iterable[str | Path], *, compself: bool, comppair: bool, verify_times: bool, num_cpus: int, **kwargs)

Load all tables and write comparison graphs for each pair.

Parameters:

input_path (Iterable[str or Path]) – Paths to input files or directories to search for tables.
compself (bool) – If True, compare each table with itself.
comppair (bool) – If True, compare every pair of different tables with the same reference and region.
verify_times (bool) – Whether to verify file modification times when loading tables.
num_cpus (int) – Number of CPUs for parallel graph writing.
**kwargs – Forwarded to each writer’s write method.

Returns:

Paths of all written output files.

Return type:

list[Path]

class seismicrna.graph.twotable.TwoTableWriter(table1: Table, table2: Table, **kwargs)

Bases: TableWriter, ABC

Write the proper types of graphs for two given tables.

abstractmethod classmethod get_graph_type(*args, **kwargs) → type[TwoTableGraph]: Type of graph.

property table1: The first table providing the data for the graph(s).

property table2: The second table providing the data for the graph(s).

seismicrna.graph.twotable.iter_table_pairs(tables: Iterable[Table]): Yield every pair of tables whose reference and region match.