smftools 0.2.3__py3-none-any.whl → 0.2.4__py3-none-any.whl

smftools/preprocessing/binarize_on_Youden.py

@@ -1,4 +1,6 @@
-def binarize_on_Youden(adata, obs_column='Reference', output_layer_name='binarized_methylation'):
+def binarize_on_Youden(adata, 
+                       ref_column='Reference_strand', 
+                       output_layer_name='binarized_methylation'):
     """
     Binarize SMF values based on position thresholds determined by calculate_position_Youden.
 
@@ -16,18 +18,18 @@ def binarize_on_Youden(adata, obs_column='Reference', output_layer_name='binariz
     binarized_methylation = np.full_like(adata.X, np.nan, dtype=float)  # Keeps same shape as adata.X
 
     # Get unique categories
-    categories = adata.obs[obs_column].cat.categories
+    references = adata.obs[ref_column].cat.categories
 
-    for cat in categories:
+    for ref in references:
         # Select subset for this category
-        cat_mask = adata.obs[obs_column] == cat
-        cat_subset = adata[cat_mask]
+        ref_mask = adata.obs[ref_column] == ref
+        ref_subset = adata[ref_mask]
 
         # Extract the probability matrix
-        original_matrix = cat_subset.X.copy()
+        original_matrix = ref_subset.X.copy()
 
         # Extract the thresholds for each position efficiently
-        thresholds = np.array(cat_subset.var[f'{cat}_position_methylation_thresholding_Youden_stats'].apply(lambda x: x[0]))
+        thresholds = np.array(ref_subset.var[f'{ref}_position_methylation_thresholding_Youden_stats'].apply(lambda x: x[0]))
 
         # Identify NaN values
         nan_mask = np.isnan(original_matrix)
@@ -39,7 +41,7 @@ def binarize_on_Youden(adata, obs_column='Reference', output_layer_name='binariz
         binarized_matrix[nan_mask] = np.nan
 
         # Assign the binarized values back into the preallocated storage
-        binarized_methylation[cat_mask, :] = binarized_matrix
+        binarized_methylation[ref_subset, :] = binarized_matrix
 
     # Store the binarized matrix in a new layer
     adata.layers[output_layer_name] = binarized_methylation

smftools/preprocessing/calculate_complexity_II.py

@@ -11,7 +11,7 @@ def calculate_complexity_II(
     n_depths=12,
     random_state=0,
     csv_summary=True,
-    uns_flag='complexity_analysis_complete',
+    uns_flag='calculate_complexity_II_performed',
     force_redo=False,
     bypass=False
 ):

smftools/preprocessing/calculate_coverage.py

@@ -1,4 +1,7 @@
-def calculate_coverage(adata, obs_column='Reference_strand', position_nan_threshold=0.00001, uns_flag='positional_coverage_calculated'):
+def calculate_coverage(adata, 
+                       ref_column='Reference_strand', 
+                       position_nan_threshold=0.01, 
+                       uns_flag='calculate_coverage_performed'):
     """
     Append position-level metadata regarding whether the position is informative within the given observation category.
 
@@ -20,32 +23,32 @@ def calculate_coverage(adata, obs_column='Reference_strand', position_nan_thresh
         # QC already performed; nothing to do
         return
     
-    categories = adata.obs[obs_column].cat.categories
+    references = adata.obs[ref_column].cat.categories
     n_categories_with_position = np.zeros(adata.shape[1])
 
-    # Loop over categories
-    for cat in categories:
-        print(f'Assessing positional coverage across samples for {cat} reference')
+    # Loop over references
+    for ref in references:
+        print(f'Assessing positional coverage across samples for {ref} reference')
 
         # Subset to current category
-        cat_mask = adata.obs[obs_column] == cat
-        temp_cat_adata = adata[cat_mask]
+        ref_mask = adata.obs[ref_column] == ref
+        temp_ref_adata = adata[ref_mask]
 
         # Compute fraction of valid coverage
-        cat_valid_coverage = np.sum(~np.isnan(temp_cat_adata.X), axis=0)
-        cat_valid_fraction = cat_valid_coverage / temp_cat_adata.shape[0]  # Avoid extra computation
+        ref_valid_coverage = np.sum(~np.isnan(temp_ref_adata.X), axis=0)
+        ref_valid_fraction = ref_valid_coverage / temp_ref_adata.shape[0]  # Avoid extra computation
 
         # Store coverage stats
-        adata.var[f'{cat}_valid_fraction'] = pd.Series(cat_valid_fraction, index=adata.var.index)
+        adata.var[f'{ref}_valid_fraction'] = pd.Series(ref_valid_fraction, index=adata.var.index)
 
         # Assign whether the position is covered based on threshold
-        adata.var[f'position_in_{cat}'] = cat_valid_fraction >= position_nan_threshold
+        adata.var[f'position_in_{ref}'] = ref_valid_fraction >= position_nan_threshold
 
         # Sum the number of categories covering each position
-        n_categories_with_position += adata.var[f'position_in_{cat}'].values
+        n_categories_with_position += adata.var[f'position_in_{ref}'].values
 
     # Store final category count
-    adata.var[f'N_{obs_column}_with_position'] = n_categories_with_position.astype(int)
+    adata.var[f'N_{ref_column}_with_position'] = n_categories_with_position.astype(int)
 
     # mark as done
     adata.uns[uns_flag] = True

smftools/preprocessing/calculate_position_Youden.py

@@ -1,7 +1,15 @@
 ## calculate_position_Youden
-
 ## Calculating and applying position level thresholds for methylation calls to binarize the SMF data
-def calculate_position_Youden(adata, positive_control_sample='positive', negative_control_sample='negative', J_threshold=0.5, obs_column='Reference', infer_on_percentile=False, inference_variable='', save=False, output_directory=''):
+def calculate_position_Youden(adata, 
+                              positive_control_sample=None, 
+                              negative_control_sample=None, 
+                              J_threshold=0.5, 
+                              ref_column='Reference_strand', 
+                              sample_column='Sample_names',
+                              infer_on_percentile=True, 
+                              inference_variable='Raw_modification_signal', 
+                              save=False, 
+                              output_directory=''):
     """
     Adds new variable metadata to each position indicating whether the position provides reliable SMF methylation calls. Also outputs plots of the positional ROC curves.
 
@@ -26,28 +34,36 @@ def calculate_position_Youden(adata, positive_control_sample='positive', negativ
     from sklearn.metrics import roc_curve, roc_auc_score
 
     control_samples = [positive_control_sample, negative_control_sample]
-    categories = adata.obs[obs_column].cat.categories 
+    references = adata.obs[ref_column].cat.categories 
     # Iterate over each category in the specified obs_column
-    for cat in categories:
-        print(f"Calculating position Youden statistics for {cat}")
+    for ref in references:
+        print(f"Calculating position Youden statistics for {ref}")
         # Subset to keep only reads associated with the category
-        cat_subset = adata[adata.obs[obs_column] == cat]
+        ref_subset = adata[adata.obs[ref_column] == ref]
         # Iterate over positive and negative control samples
-        for control in control_samples:
+        for i, control in enumerate(control_samples):
             # Initialize a dictionary for the given control sample. This will be keyed by dataset and position to point to a tuple of coordinate position and an array of methylation probabilities
-            adata.uns[f'{cat}_position_methylation_dict_{control}'] = {}
-            if infer_on_percentile:
-                sorted_column = cat_subset.obs[inference_variable].sort_values(ascending=False)
-                if control == "positive":
+            adata.uns[f'{ref}_position_methylation_dict_{control}'] = {}
+            # If controls are not passed and infer on percentile is True, infer thresholds based on top and bottom percentile windows for a given obs column metric.
+            if infer_on_percentile and not control:
+                sorted_column = ref_subset.obs[inference_variable].sort_values(ascending=False)
+                if i == 0:
+                    control == 'positive'
+                    positive_control_sample = control
                     threshold = np.percentile(sorted_column, 100 - infer_on_percentile)
-                    control_subset = cat_subset[cat_subset.obs[inference_variable] >= threshold, :]
+                    control_subset = ref_subset[ref_subset.obs[inference_variable] >= threshold, :]
                 else:
+                    control == 'negative'
+                    negative_control_sample = control
                     threshold = np.percentile(sorted_column, infer_on_percentile)
-                    control_subset = cat_subset[cat_subset.obs[inference_variable] <= threshold, :]   
+                    control_subset = ref_subset[ref_subset.obs[inference_variable] <= threshold, :]   
+            elif not infer_on_percentile and not control:
+                print("Can not threshold Anndata on Youden threshold. Need to either provide control samples or set infer_on_percentile to True")
+                return
             else:
                 # get the current control subset on the given category
-                filtered_obs = cat_subset.obs[cat_subset.obs['Sample_names'].str.contains(control, na=False, regex=True)]
-                control_subset = cat_subset[filtered_obs.index]
+                filtered_obs = ref_subset.obs[ref_subset.obs[sample_column] == control]
+                control_subset = ref_subset[filtered_obs.index]
             # Iterate through every position in the control subset
             for position in range(control_subset.shape[1]):
                 # Get the coordinate name associated with that position
@@ -63,9 +79,9 @@ def calculate_position_Youden(adata, positive_control_sample='positive', negativ
                 # Get fraction coverage
                 fraction_coverage = position_coverage / control_subset.shape[0]
                 # Save the position and the position methylation data for the control subset
-                adata.uns[f'{cat}_position_methylation_dict_{control}'][f'{position}'] = (position, position_data, fraction_coverage)
+                adata.uns[f'{ref}_position_methylation_dict_{control}'][f'{position}'] = (position, position_data, fraction_coverage)
 
-    for cat in categories:
+    for ref in references:
         fig, ax = plt.subplots(figsize=(6, 4))
         plt.plot([0, 1], [0, 1], linestyle='--', color='gray')
         plt.xlabel('False Positive Rate')
@@ -76,13 +92,13 @@ def calculate_position_Youden(adata, positive_control_sample='positive', negativ
         n_total_positions = 0
         # Initialize a list that will hold the positional thresholds for the category
         probability_thresholding_list = [(np.nan, np.nan)] * adata.shape[1]
-        for i, key in enumerate(adata.uns[f'{cat}_position_methylation_dict_{positive_control_sample}'].keys()):
-            position = int(adata.uns[f'{cat}_position_methylation_dict_{positive_control_sample}'][key][0])
-            positive_position_array = adata.uns[f'{cat}_position_methylation_dict_{positive_control_sample}'][key][1]
-            fraction_coverage = adata.uns[f'{cat}_position_methylation_dict_{positive_control_sample}'][key][2]
+        for i, key in enumerate(adata.uns[f'{ref}_position_methylation_dict_{positive_control_sample}'].keys()):
+            position = int(adata.uns[f'{ref}_position_methylation_dict_{positive_control_sample}'][key][0])
+            positive_position_array = adata.uns[f'{ref}_position_methylation_dict_{positive_control_sample}'][key][1]
+            fraction_coverage = adata.uns[f'{ref}_position_methylation_dict_{positive_control_sample}'][key][2]
             if fraction_coverage > 0.2:
                 try:
-                    negative_position_array = adata.uns[f'{cat}_position_methylation_dict_{negative_control_sample}'][key][1] 
+                    negative_position_array = adata.uns[f'{ref}_position_methylation_dict_{negative_control_sample}'][key][1] 
                     # Combine the negative and positive control data
                     data = np.concatenate([negative_position_array, positive_position_array])
                     labels = np.array([0] * len(negative_position_array) + [1] * len(positive_position_array))
@@ -101,7 +117,7 @@ def calculate_position_Youden(adata, positive_control_sample='positive', negativ
                         plt.plot(fpr, tpr, label='ROC curve')
                 except:
                     probability_thresholding_list[position] = (0.8, np.nan)
-        title = f'ROC Curve for {n_passed_positions} positions with J-stat greater than {J_threshold}\n out of {n_total_positions} total positions on {cat}'
+        title = f'ROC Curve for {n_passed_positions} positions with J-stat greater than {J_threshold}\n out of {n_total_positions} total positions on {ref}'
         plt.title(title)
         save_name = output_directory / f"{title}.png"
         if save:
@@ -110,6 +126,6 @@ def calculate_position_Youden(adata, positive_control_sample='positive', negativ
         else:
             plt.show()   
 
-        adata.var[f'{cat}_position_methylation_thresholding_Youden_stats'] = probability_thresholding_list
+        adata.var[f'{ref}_position_methylation_thresholding_Youden_stats'] = probability_thresholding_list
         J_max_list = [probability_thresholding_list[i][1] for i in range(adata.shape[1])]
-        adata.var[f'{cat}_position_passed_QC'] = [True if i > J_threshold else False for i in J_max_list]
+        adata.var[f'{ref}_position_passed_QC'] = [True if i > J_threshold else False for i in J_max_list]

smftools/preprocessing/calculate_read_modification_stats.py

@@ -2,7 +2,7 @@ def calculate_read_modification_stats(adata,
                                       reference_column, 
                                       sample_names_col, 
                                       mod_target_bases,
-                                      uns_flag="read_modification_stats_calculated",
+                                      uns_flag="calculate_read_modification_stats_performed",
                                       bypass=False,
                                       force_redo=False
 ):

smftools/preprocessing/filter_reads_on_length_quality_mapping.py

@@ -11,7 +11,7 @@ def filter_reads_on_length_quality_mapping(
     length_ratio: Optional[Sequence[float]] = None,         # e.g. [min, max]
     read_quality: Optional[Sequence[float]] = None,         # e.g. [min, max]  (commonly min only)
     mapping_quality: Optional[Sequence[float]] = None,      # e.g. [min, max]  (commonly min only)
-    uns_flag: str = "reads_removed_failing_length_quality_mapping_qc",
+    uns_flag: str = "filter_reads_on_length_quality_mapping_performed",
     bypass: bool = False,
     force_redo: bool = True
 ) -> ad.AnnData:

smftools/preprocessing/filter_reads_on_modification_thresholds.py

@@ -15,7 +15,7 @@ def filter_reads_on_modification_thresholds(
     a_thresholds: Optional[Sequence[float]] = None,
     use_other_c_as_background: bool = False,
     min_valid_fraction_positions_in_read_vs_ref: Optional[float] = None,
-    uns_flag: str = 'reads_filtered_on_modification_thresholds',
+    uns_flag: str = 'filter_reads_on_modification_thresholds_performed',
     bypass: bool = False,
     force_redo: bool = False,
     reference_column: str = 'Reference_strand',

smftools/preprocessing/flag_duplicate_reads.py

@@ -77,7 +77,7 @@ def flag_duplicate_reads(
     sample_col: str = "Barcode",
     output_directory: Optional[str] = None,
     metric_keys: Union[str, List[str]] = ("Fraction_any_C_site_modified",),
-    uns_flag: str = "read_duplicate_detection_performed",
+    uns_flag: str = "flag_duplicate_reads_performed",
     uns_filtered_flag: str = "read_duplicates_removed",
     bypass: bool = False,
     force_redo: bool = False,

smftools/preprocessing/invert_adata.py

@@ -1,6 +1,6 @@
 ## invert_adata
 
-def invert_adata(adata, uns_flag='adata_positions_inverted', force_redo=False):
+def invert_adata(adata, uns_flag='invert_adata_performed', force_redo=False):
     """
     Inverts the AnnData object along the column (variable) axis.
 

smftools/preprocessing/load_sample_sheet.py

@@ -2,7 +2,7 @@ def load_sample_sheet(adata,
                       sample_sheet_path, 
                       mapping_key_column='obs_names',
                       as_category=True,
-                      uns_flag='sample_sheet_loaded',
+                      uns_flag='load_sample_sheet_performed',
                       force_reload=True
                         ):
     """

smftools/preprocessing/reindex_references_adata.py

@@ -0,0 +1,37 @@
+## reindex_references_adata
+
+def reindex_references_adata(adata, 
+                             reference_col="Reference_strand", 
+                             offsets=None, 
+                             new_col="reindexed",
+                             uns_flag='reindex_references_adata_performed', 
+                             force_redo=False):
+    
+    # Only run if not already performed
+    already = bool(adata.uns.get(uns_flag, False))
+    if (already and not force_redo):
+        return None
+    
+    if offsets is None:
+        pass
+    else:
+        # Ensure var_names are numeric
+        var_coords = adata.var_names.astype(int)
+
+        for ref in adata.obs[reference_col].unique():
+            if ref not in offsets:
+                pass
+            else:
+                offset_value = offsets[ref]
+
+                # Create a new var column for this reference
+                colname = f"{ref}_{new_col}"
+
+                # Add offset to all var positions
+                adata.var[colname] = var_coords + offset_value
+
+    # mark as done
+    adata.uns[uns_flag] = True
+
+    print("Reindexing complete!")
+    return None

smftools/readwrite.py

@@ -722,6 +722,100 @@ def safe_write_h5ad(adata, path, compression="gzip", backup=False, backup_dir=No
             print(" -", e)
 
     print("=== end report ===\n")
+
+    # ---------- create CSV output directory ----------
+    try:
+        csv_dir = path.parent / "csvs"
+        csv_dir.mkdir(exist_ok=True)
+        if verbose:
+            print(f"CSV outputs will be written to: {csv_dir}")
+    except Exception as e:
+        msg = f"Failed to create CSV output directory: {e}"
+        report['errors'].append(msg)
+        if verbose:
+            print(msg)
+        csv_dir = path.parent  # fallback just in case
+
+    # ---------- write keys summary CSV ----------
+    try:
+        meta_rows = []
+
+        # obs columns
+        for col in adata_copy.obs.columns:
+            meta_rows.append({
+                "kind": "obs",
+                "name": col,
+                "dtype": str(adata_copy.obs[col].dtype),
+            })
+
+        # var columns
+        for col in adata_copy.var.columns:
+            meta_rows.append({
+                "kind": "var",
+                "name": col,
+                "dtype": str(adata_copy.var[col].dtype),
+            })
+
+        # layers
+        for k, v in adata_copy.layers.items():
+            meta_rows.append({
+                "kind": "layer",
+                "name": k,
+                "dtype": str(np.asarray(v).dtype),
+            })
+
+        # obsm
+        for k, v in adata_copy.obsm.items():
+            meta_rows.append({
+                "kind": "obsm",
+                "name": k,
+                "dtype": str(np.asarray(v).dtype),
+            })
+
+        # uns
+        for k, v in adata_copy.uns.items():
+            meta_rows.append({
+                "kind": "uns",
+                "name": k,
+                "dtype": type(v).__name__,
+            })
+
+        meta_df = pd.DataFrame(meta_rows)
+
+        # same base name, inside csvs/
+        base = path.stem    # removes .h5ad
+        meta_path = csv_dir / f"{base}.keys.csv"
+
+        meta_df.to_csv(meta_path, index=False)
+        if verbose:
+            print(f"Wrote keys summary CSV to {meta_path}")
+
+    except Exception as e:
+        msg = f"Failed to write keys CSV: {e}"
+        report["errors"].append(msg)
+        if verbose:
+            print(msg)
+
+    # ---------- write full obs and var dataframes ----------
+    try:
+        base = path.stem
+
+        obs_path = csv_dir / f"{base}.obs.csv"
+        var_path = csv_dir / f"{base}.var.csv"
+
+        adata_copy.obs.to_csv(obs_path, index=True)
+        adata_copy.var.to_csv(var_path, index=True)
+
+        if verbose:
+            print(f"Wrote obs DataFrame to {obs_path}")
+            print(f"Wrote var DataFrame to {var_path}")
+
+    except Exception as e:
+        msg = f"Failed to write obs/var CSVs: {e}"
+        report["errors"].append(msg)
+        if verbose:
+            print(msg)
+
     return report
 
 def safe_read_h5ad(path, backup_dir=None, restore_backups=True, re_categorize=True, categorical_threshold=100, verbose=True):

{smftools-0.2.3.dist-info → smftools-0.2.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: smftools
-Version: 0.2.3
+Version: 0.2.4
 Summary: Single Molecule Footprinting Analysis in Python.
 Project-URL: Source, https://github.com/jkmckenna/smftools
 Project-URL: Documentation, https://smftools.readthedocs.io/
@@ -96,30 +96,36 @@ Description-Content-Type: text/markdown
 [![Docs](https://readthedocs.org/projects/smftools/badge/?version=latest)](https://smftools.readthedocs.io/en/latest/?badge=latest)
 
 # smftools
-A Python tool for processing raw sequencing data derived from single molecule footprinting experiments into [anndata](https://anndata.readthedocs.io/en/latest/) objects. Additional functionality for preprocessing, analysis, and visualization.
+A Python tool for processing raw sequencing data derived from single molecule footprinting experiments into [anndata](https://anndata.readthedocs.io/en/latest/) objects. Additional functionality for preprocessing, spatial analyses, and HMM based feature annotation.
 
 ## Philosophy
-While most genomic data structures handle low-coverage data (<100X) along large references, smftools prioritizes high-coverage data (scalable to >1,000,000X coverage) of a few genomic loci at a time. This enables efficient data storage, rapid data operations, hierarchical metadata handling, seamless integration with various machine-learning packages, and ease of visualization. Furthermore, functionality is modularized, enabling analysis sessions to be saved, reloaded, and easily shared with collaborators. Analyses are centered around the [anndata](https://anndata.readthedocs.io/en/latest/) object, and are heavily inspired by the work conducted within the single-cell genomics community.
+While genomic data structures (SAM/BAM) were built to handle low-coverage data (<1000X) along large references, smftools prioritizes high-coverage data (scalable to >1,000,000X coverage) of a few genomic loci at a time. This enables efficient data storage, rapid data operations, hierarchical metadata handling, seamless integration with various machine-learning packages, and ease of visualization. Furthermore, functionality is modularized, enabling analysis sessions to be saved, reloaded, and easily shared with collaborators. Analyses are centered around the [anndata](https://anndata.readthedocs.io/en/latest/) object, and are heavily inspired by the work conducted within the single-cell genomics community.
 
 ## Dependencies
 The following CLI tools need to be installed and configured before using the informatics (smftools.inform) module of smftools:
 1) [Dorado](https://github.com/nanoporetech/dorado) -> Basecalling, alignment, demultiplexing.
 2) [Minimap2](https://github.com/lh3/minimap2) -> Alignment if not using dorado.
-3) [Modkit](https://github.com/nanoporetech/modkit) -> Extracting read level methylation metrics from modified BAM files.
+3) [Modkit](https://github.com/nanoporetech/modkit) -> Extracting read level methylation metrics from modified BAM files. Only required for direct modification detection (ie methylation).
 
-## Modules
-### Informatics: Processes raw Nanopore/Illumina data from SMF experiments into an AnnData object.
+## Main Commands
+### smftools load: Processes raw Nanopore/Illumina data from SMF experiments into an AnnData object.
 ![](docs/source/_static/smftools_informatics_diagram.png)
-### Preprocessing: Appends QC metrics to the AnnData object and performs filtering.
+### smftools preprocess: Appends QC metrics to the AnnData object and performs filtering.
 ![](docs/source/_static/smftools_preprocessing_diagram.png)
-### Tools: Appends analyses to the AnnData object.
-- Currently Includes: Position X Position correlation matrices, Hidden Markov Model feature detection, clustering, dimensionality reduction, peak calling, train/test workflows for various ML classifiers.
-- To do: Additional ML methods for learning predictive single molecule features on condition labels: Autoencoders, Variational Autoencoders, Transformers.
-### Plotting: Visualization of analyses stored within the AnnData object.
-- Most analyses appended to the adata object by a tools method have, or will have, an accompanying plotting method.
+### smftools spatial: Appends spatial analyses to the AnnData object.
+- Currently Includes: Position X Position correlation matrices, clustering, dimensionality reduction, spatial autocorrelation. 
+### smftools hmm: Fits a basic HMM to each sample and appends HMM feature layers
+- Main outputs wills be stored in adata.layers
+### smftools batch <command>: Performs batch processing on a csv of config file pathes for any of the above commands.
+- Nice when analyzing multiple experiments
+### smftools concatenate: Concatenates a list or directory of anndata objects.
+- Mainly used for combining multiple experiments into a single anndata object.
 
 ## Announcements
 
+### 12/02/25 - Version 0.2.3 is available through PyPI
+Version 0.2.3 provides the core smftools functionality through several command line commands (load, preprocess, spatial, hmm).
+
 ### 11/05/25 - Version 0.2.1 is available through PyPI
 Version 0.2.1 makes the core workflow (smftools load) a command line tool that takes in an experiment_config.csv file for input/output and parameter management.
 

{smftools-0.2.3.dist-info → smftools-0.2.4.dist-info}/RECORD

@@ -1,44 +1,46 @@
 smftools/__init__.py,sha256=aZlrZBVexf_nEnzQeZu7NU_Kp6OnxcYpLo1KPImi7sI,599
 smftools/_settings.py,sha256=Ed8lzKUA5ncq5ZRfSp0t6_rphEEjMxts6guttwTZP5Y,409
-smftools/_version.py,sha256=X0PliCRFAeVnSTceUeHX1eM0j1HFhGFDWCRxLdde2Bs,21
-smftools/cli_entry.py,sha256=_QdtEKcVK5o-e5s9ETB9sOIdftPVlrDxvvjBKcP6YNk,14680
-smftools/readwrite.py,sha256=ExKZHNZ0QB-PtSck08drXfHTqbPeSUTHiYhv951SH1s,45994
+smftools/_version.py,sha256=k2uKAAzDEmm1BIVWeztFlHrCh9fq64H6szFcsXW7tvs,21
+smftools/cli_entry.py,sha256=LvobMVtEb_jrLZScoWCB-OBjUMue9JQBXJZW1oMbHnw,14618
+smftools/readwrite.py,sha256=mbuCKj7LfEKp4bDBxxxMiaTddMwblwURpcCKpgmU6Sw,48678
 smftools/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-smftools/cli/cli_flows.py,sha256=xRiFUThoAL3LX1xdXaHVg4LjyJI4uNpGsc9aQ_wVCto,4941
-smftools/cli/hmm_adata.py,sha256=PApUJW0lO4kcLjsiqqQopXgL3Dg-AascIqJrgvSY1Rg,15916
-smftools/cli/load_adata.py,sha256=Qt1ej-osyJ47fpBkGaSDgR1F8E4aBNAdcXeBAGM-Lqg,29100
-smftools/cli/preprocess_adata.py,sha256=EKGbSTli7qvL44OQUmMalYJjsH9vn3w4Rx7U7BL0ybs,20991
-smftools/cli/spatial_adata.py,sha256=AX6iyBfbXud9actteTvDuaQUU_SE3SyBIeknR317g34,30212
+smftools/cli/helpers.py,sha256=tgjxUlOIhFGCLGD2ON7zlD45UPx93vENM82mM_BpLFk,1281
+smftools/cli/hmm_adata.py,sha256=2ria8u6cCBQnzX_GjUUO3wBVOd7a4m3Al-vzwk0OasQ,17728
+smftools/cli/load_adata.py,sha256=W4NgbM28wOzQHkLnZNILJyblRgee-O4oLnNZcyPDCXc,30486
+smftools/cli/preprocess_adata.py,sha256=g9aHQ1DSScb4zx8RfpCjcEmam6APWHiu8Ow0sza6D2Y,22203
+smftools/cli/spatial_adata.py,sha256=pp0KLK8d-MYjl_hF1ziDVKc6uOJGDDDbKNQELQcRUa8,28980
+smftools/cli/archived/cli_flows.py,sha256=xRiFUThoAL3LX1xdXaHVg4LjyJI4uNpGsc9aQ_wVCto,4941
 smftools/config/__init__.py,sha256=ObUnnR7aRSoD_uvpmsxA_BUFt4NOOfWNopDVCqjp7tg,69
-smftools/config/conversion.yaml,sha256=HrFz2f9QRe1RuhmgU6ZtMHaM4ZzY61_aLcugsmpV40Q,969
-smftools/config/deaminase.yaml,sha256=mw2aY222y2xg08Rs5CWvjlrXo3vaEim7JwBThA80y4o,1349
-smftools/config/default.yaml,sha256=3IrX0OrUyjhVc3CqTjM8uiprKWrrBdVtil4YhtVzKdQ,10233
-smftools/config/direct.yaml,sha256=SBhdtG7PKm-z5xxQmA7JV3NQsGnUJ4p58fGH8BnoMrM,2137
+smftools/config/conversion.yaml,sha256=07dKEXykQeP5VoVxa4xst-tcbSX4B6ErqyqtWJ5RCKk,1177
+smftools/config/deaminase.yaml,sha256=okXdMFAghUAsDyx6P5Kru7ydF2bcbrhMPOaMpXlZPGM,1359
+smftools/config/default.yaml,sha256=cKUUxVkH42kkHQM82mNJC8bfcak6lY063AnIif5o-1g,13071
+smftools/config/direct.yaml,sha256=s30JbOTOOdIiBIefPSEi72YABHnfcCyFXj9WwZ7duJQ,2173
 smftools/config/discover_input_files.py,sha256=G9vyAmK_n_8Ur5dOnumevVLG3ydHchMy_JQrJdiuuz0,3892
-smftools/config/experiment_config.py,sha256=d_6f_Uv3CY-1orHbxpHtAZDsY2gwxw079_pNgR9wDUg,58837
+smftools/config/experiment_config.py,sha256=f7hVIc9ShUZk852Ypp6Dfelus8iKFHrSbThiyhpuQsE,63259
 smftools/datasets/F1_hybrid_NKG2A_enhander_promoter_GpC_conversion_SMF.h5ad.gz,sha256=q6wJtgFRDln0o20XNCx1qad3lwcdCoylqPN7wskTfI8,2926497
 smftools/datasets/F1_sample_sheet.csv,sha256=9PodIIOXK2eamYPbC6DGnXdzgi9bRDovf296j1aM0ak,259
 smftools/datasets/__init__.py,sha256=xkSTlPuakVYVCuRurif9BceNBDt6bsngJvvjI8757QI,142
 smftools/datasets/dCas9_m6A_invitro_kinetics.h5ad.gz,sha256=niOcVHaYY7h3XyvwSkN-V_NMBaRt2vTP5TrJO0CwMCs,8385050
 smftools/datasets/datasets.py,sha256=0y597Ntp707bOgDwN6O-JEt9yxgplj66p0aj6Zs_IB4,779
-smftools/hmm/HMM.py,sha256=K8rt-EHn3ylIHpQ3dHf_OZCXxCBVSS2UWTgSGOatwHw,71046
-smftools/hmm/__init__.py,sha256=BkX145eGVy-kFOtyqOcu-Hzv9ZJLDQ3cfDe51eKBTwY,585
-smftools/hmm/apply_hmm_batched.py,sha256=BBeJ8DiIuuMWzLwtDdk2DO2vvrfLCrVe4JtRYPFItIU,10648
-smftools/hmm/calculate_distances.py,sha256=KDWimQ6u-coyxCKrbTm42Fh_Alf_gURBZ0vfFaem848,644
-smftools/hmm/call_hmm_peaks.py,sha256=T-3Ld8H4t3Mgg2whBTYP9s2QL7rY-9RIzVCgB6avKhE,4625
+smftools/hmm/HMM.py,sha256=Y7YB-45HoLN--JloajoLBgC0rIYmHuWHDfmKRXfFuFk,71458
+smftools/hmm/__init__.py,sha256=_-plMbL5xq8d0szNIYgUrgUwdb8oybuyTn6jned8eSU,382
+smftools/hmm/call_hmm_peaks.py,sha256=BMlwDh-_k8bzqRn4LSYuTk3dCcUoNYHp8eohvWYNn7A,14573
 smftools/hmm/display_hmm.py,sha256=3WuQCPvM3wPfzAdgbhfiBTd0g5mQdx9HTUdqAxs2aj4,825
 smftools/hmm/hmm_readwrite.py,sha256=DjJ3hunpBQ7N0GVvxL7-0QUas_SkA88LVgL72mVK2cI,359
 smftools/hmm/nucleosome_hmm_refinement.py,sha256=nQWimvse6dclcXhbU707rGbRVMKHM0mU_ZhH9g2yCMA,4641
-smftools/hmm/train_hmm.py,sha256=srzRcB9LEmNuHyBM0R5Z0VEnxecifQt-MoaJhADxGT8,2477
+smftools/hmm/archived/apply_hmm_batched.py,sha256=BBeJ8DiIuuMWzLwtDdk2DO2vvrfLCrVe4JtRYPFItIU,10648
+smftools/hmm/archived/calculate_distances.py,sha256=KDWimQ6u-coyxCKrbTm42Fh_Alf_gURBZ0vfFaem848,644
+smftools/hmm/archived/call_hmm_peaks.py,sha256=T-3Ld8H4t3Mgg2whBTYP9s2QL7rY-9RIzVCgB6avKhE,4625
+smftools/hmm/archived/train_hmm.py,sha256=srzRcB9LEmNuHyBM0R5Z0VEnxecifQt-MoaJhADxGT8,2477
 smftools/informatics/__init__.py,sha256=vLvSrCtCVYRUCCNLW7fL3ltPr3h_w8FhT--V6el3ZkQ,1191
-smftools/informatics/bam_functions.py,sha256=otgl3TRPLn5Fnsx1jXX75du90k3XB3RHGzlfamvETsU,32670
+smftools/informatics/bam_functions.py,sha256=SCtOQWgF7Nqbk7-22fAq9J8kRYrd2V5chmM0x1lLJh0,32261
 smftools/informatics/basecalling.py,sha256=jc39jneaa8Gt1azutHgBGWHqCoPeTVSGBu3kyQwP7xM,3460
 smftools/informatics/bed_functions.py,sha256=uETVxT5mRWDNn7t0OqhDi8kDiq7uDakeHB1L2JsP4PA,13377
 smftools/informatics/binarize_converted_base_identities.py,sha256=yOepGaNBGfZJEsMiLRwKauvsmaHn_JRrxaGp8LmKAXs,7778
 smftools/informatics/complement_base_list.py,sha256=k6EkLtxFoajaIufxw1p0pShJ2nPHyGLTbzZmIFFjB4o,532
 smftools/informatics/converted_BAM_to_adata.py,sha256=Y2kQNWly0WjjGN9El9zL1nLfjVxmPLWONvX5VNgZUh0,22554
 smftools/informatics/fasta_functions.py,sha256=5IfTkX_GIj5gRJB9PjL_WjyEktpBHwGsmS_nnO1ETjI,9790
-smftools/informatics/h5ad_functions.py,sha256=iAOxJjhaDslTUC78kjUHlCELigDl73sWo0fvXcKuFoI,7824
+smftools/informatics/h5ad_functions.py,sha256=9zUKuARwjjt0J-i_kBqo2jxLtD6Gud1VxKT0pV-ACeA,7829
 smftools/informatics/modkit_extract_to_adata.py,sha256=TrgrL_IgfqzNJ9qZ_2EvF_B38_Syw8mP38Sl7v0Riwo,55278
 smftools/informatics/modkit_functions.py,sha256=lywjeqAJ7Cdd7k-0P3YaL_9cAZvEDTDLh91rIRcSMWE,5604
 smftools/informatics/ohe.py,sha256=MEmh3ps-ZSSyXuIrr5LMzQvCsDJRCYiy7JS-WD4TlYs,5805
@@ -118,37 +120,38 @@ smftools/machine_learning/utils/grl.py,sha256=BWBDp_kQBigrUzQpRbZzgpfr_WOcd2K2V3
 smftools/plotting/__init__.py,sha256=7T3-hZFgTY0nfQgV4J6Vn9ogwkNMlY315kguZR7V1AI,866
 smftools/plotting/autocorrelation_plotting.py,sha256=cF9X3CgKiwzL79mgMUFO1tSqdybDoPN1COQQ567InCY,27455
 smftools/plotting/classifiers.py,sha256=8_zabh4NNB1_yVxLD22lfrfl5yfzbEoG3XWqlIqdtrQ,13786
-smftools/plotting/general_plotting.py,sha256=2JzE7agm_tILpQ67BHs5pdyPRsHBwcENZe7n4gfMWgM,61350
+smftools/plotting/general_plotting.py,sha256=o4aPXm_2JRj69XyHINKSTAJGaw9VA-csDgX1pyirso0,63151
 smftools/plotting/hmm_plotting.py,sha256=3Eq82gty_0b8GkSMCQgUlbKfzR9h2fJ5rZkB8yYGX-M,10934
-smftools/plotting/position_stats.py,sha256=4XukYIWeWZ_aGSZg1K0t37KA2aknjNNKT5kcKFfuz8Q,17428
+smftools/plotting/position_stats.py,sha256=Ia15EuYq5r3Ckz3jVjYMHON6IHZboatAVqJdb2WrUA4,17415
 smftools/plotting/qc_plotting.py,sha256=q5Ri0q89udvNUFUNxHzgk9atvQYqUkqkS5-JFq9EqoI,10045
-smftools/preprocessing/__init__.py,sha256=GAQBULUH7fGVabzK5Cq5Wj-0ew0vNA-jWQtR5LAowvs,1746
-smftools/preprocessing/add_read_length_and_mapping_qc.py,sha256=zD_Kxw3DvyOypfuSMGv0ESyt-02w4XlAAMqQxb7yDNQ,5700
-smftools/preprocessing/append_base_context.py,sha256=wGBAADePnys8DLUR15MpRe2BUcfCMDJWaCDDNyjn6AU,6209
-smftools/preprocessing/append_binary_layer_by_base_context.py,sha256=s-7t-VKCs9Y67pX7kH6DNCEkC-RW4nM-UPsBQV2ZwtE,6186
+smftools/preprocessing/__init__.py,sha256=mcmovdFq6jt1kWIe0sVW6MwCXs4tUVTy3Qak7RDts74,1644
+smftools/preprocessing/append_base_context.py,sha256=VnxKf8sI4uWale215FEFFoE2me6uJszXvswl-dFQmUY,6702
+smftools/preprocessing/append_binary_layer_by_base_context.py,sha256=qgjeDyfOghuqWZAzCjd4eE5riCWAgra6CIZ9UCyUgTs,6207
 smftools/preprocessing/binarize.py,sha256=6Vr7Z8zgtJ5rS_uPAx1n3EnQR670V33DlZ_95JmOeWc,484
-smftools/preprocessing/binarize_on_Youden.py,sha256=HGs4p7XiOSYU3_z8QswNHIA9HlrI-7Pp1Kggrn6yUnI,1834
+smftools/preprocessing/binarize_on_Youden.py,sha256=OwI0JwKBsSPVdPr61D31dR9XhnF0N4e5PnbboTpk8xI,1891
 smftools/preprocessing/binary_layers_to_ohe.py,sha256=Lxd8knelNTaUozfGMFNMlnrOb6uP28Laj3Ymw6cRHL0,1826
-smftools/preprocessing/calculate_complexity.py,sha256=cXMpFrhkwkPipQo2GZGT5yFknMYUMt1t8gz0Cse1DrA,3288
-smftools/preprocessing/calculate_complexity_II.py,sha256=DGfl0jkuBPUpzhKVItN0W7EPzh-QYuR4IxRObPE6gAQ,9301
+smftools/preprocessing/calculate_complexity_II.py,sha256=oh5y0jbM1-k29ujRUfvXoL3ir4E6bVXLE9bWxlD5efc,9306
 smftools/preprocessing/calculate_consensus.py,sha256=6zRpRmb2xdfDu5hctZrReALRb7Pjn8sy8xJZTm3o0nU,2442
-smftools/preprocessing/calculate_coverage.py,sha256=4WTILzKLzxGLSsQrZkshXP-IRQpoVu3Fkqc0QTpux3Y,2132
+smftools/preprocessing/calculate_coverage.py,sha256=L417_XWAadMH3vxVDGEEAqxIGOiV48nfzVzD7HYyhus,2199
 smftools/preprocessing/calculate_pairwise_differences.py,sha256=5zJbNNaFld5qgKRoPyplCmMHflbvAQ9eKWCXPXPpJ60,1774
 smftools/preprocessing/calculate_pairwise_hamming_distances.py,sha256=e5Mzyex7pT29H2PY014uU4Fi_eewbut1JkzC1ffBbCg,961
-smftools/preprocessing/calculate_position_Youden.py,sha256=yaSd6UDXPCddoN1UR6LgTqE5teJ79Ldw0BAlemc9fB4,7453
+smftools/preprocessing/calculate_position_Youden.py,sha256=JJLvU62zpBcvWm5QnsQ3FeRgIv5TMQbz5zTHa3z_Y1s,8342
 smftools/preprocessing/calculate_read_length_stats.py,sha256=gNNePwMqYZJidzGgT1ZkfSlvc5Y3I3bi5KNYpP6wQQc,4584
-smftools/preprocessing/calculate_read_modification_stats.py,sha256=mIlLBqNflVIkuoLxhbyujq3JEKyPl8iebhUlikB9brM,4775
+smftools/preprocessing/calculate_read_modification_stats.py,sha256=hZzoEe1Acc1TQV3crkjyGZBWTMkMMcqXymJb3vJMHks,4784
 smftools/preprocessing/clean_NaN.py,sha256=IOcnN5YF05gpPQc3cc3IS83petCnhCpkYiyT6bXEyx0,1937
 smftools/preprocessing/filter_adata_by_nan_proportion.py,sha256=GZcvr2JCsthX8EMw34S9-W3fc6JElw6ka99Jy6f2JvA,1292
-smftools/preprocessing/filter_reads_on_length_quality_mapping.py,sha256=93LgTy_vsPnOZgoiXhZ1-w_pix2oFdBk-dsBUoz33Go,7379
-smftools/preprocessing/filter_reads_on_modification_thresholds.py,sha256=4TUvChkSH8R4p_0TpRCh7TounkdUgQHh71TGNmsZ29A,19355
-smftools/preprocessing/flag_duplicate_reads.py,sha256=MySI9En6xVp0FqL7hfiLw0EP3JnGVJWM_yZfkvN-m1U,65585
-smftools/preprocessing/invert_adata.py,sha256=HYMJ1sR3Ui8j6bDjY8OcVQOETzZV-_rrpIYaWLZL6S4,1049
-smftools/preprocessing/load_sample_sheet.py,sha256=AjJf2MrqGHJJ2rNjYi09zV1QkLTq8qGaHGVklXHnPuU,1908
+smftools/preprocessing/filter_reads_on_length_quality_mapping.py,sha256=UhMXpM_qxbhTCorjpKAePRk1qQVls8DP6Z51aFVnr3k,7380
+smftools/preprocessing/filter_reads_on_modification_thresholds.py,sha256=LK3u0mIwD-T_qwqIH8v7BP1ZRL88HtRXPkDJwchsCjk,19363
+smftools/preprocessing/flag_duplicate_reads.py,sha256=8Z3sVQr8gmci3ZtYfQGDAHI7GpKGhzmAFHoZVyL6nK4,65581
+smftools/preprocessing/invert_adata.py,sha256=TmvwRGlkJKnMajOADAzpE_C2kYEtDVYDYtQKv3IthKs,1047
+smftools/preprocessing/load_sample_sheet.py,sha256=WXAKfIhbnptnkbIpI5hEe6p02HhpQ3eRX1EDGEEvH-8,1916
 smftools/preprocessing/make_dirs.py,sha256=lWHXpwC76MFM5sSme9i_WeYUaxutzybendokhny03ds,537
 smftools/preprocessing/min_non_diagonal.py,sha256=hx1asW8CEmLaIroZISW8EcAf_RnBEC_nofGD8QG0b1E,711
 smftools/preprocessing/recipes.py,sha256=cfKEpKW8TtQLe1CMdSHyPuIgKiWOPn7uP6uMIoRlnaQ,7063
+smftools/preprocessing/reindex_references_adata.py,sha256=4oViEcWWSi7bnX3Yyf-DdSZBSocvuiqr4LC-jDFHwu0,1137
 smftools/preprocessing/subsample_adata.py,sha256=ivJvJIOvEtyvAjqZ7cwEeVedm4QgJxCJEI7sFaTuI3w,2360
+smftools/preprocessing/archives/add_read_length_and_mapping_qc.py,sha256=zD_Kxw3DvyOypfuSMGv0ESyt-02w4XlAAMqQxb7yDNQ,5700
+smftools/preprocessing/archives/calculate_complexity.py,sha256=cXMpFrhkwkPipQo2GZGT5yFknMYUMt1t8gz0Cse1DrA,3288
 smftools/preprocessing/archives/mark_duplicates.py,sha256=kwfstcWb7KkqeNB321dB-NLe8yd9_hZsSmpL8pCVBQg,8747
 smftools/preprocessing/archives/preprocessing.py,sha256=4mLT09A7vwRZ78FHmuwtv38mH9TQ9qrZc_WjHRhhkIw,34379
 smftools/preprocessing/archives/remove_duplicates.py,sha256=Erooi5_1VOUNfWpzddzmMNYMCl1U1jJryt7ZtMhabAs,699
@@ -166,8 +169,8 @@ smftools/tools/archived/classify_methylated_features.py,sha256=Z0N2UKw3luD3CTQ8w
 smftools/tools/archived/classify_non_methylated_features.py,sha256=IJERTozEs7IPL7K-VIjq2q2K36wRCW9iiNSYLAXasrA,3256
 smftools/tools/archived/subset_adata_v1.py,sha256=qyU9iCal03edb5aUS3AZ2U4TlL3uQ42jGI9hX3QF7Fc,1047
 smftools/tools/archived/subset_adata_v2.py,sha256=OKZoUpvdURPtckIQxGTWmOI5jLa-_EU62Xs3LyyehnA,1880
-smftools-0.2.3.dist-info/METADATA,sha256=w_PRsBPndPoTQZviW9WTuiZV1Pk3ukeJ155OvC4E57M,8787
-smftools-0.2.3.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
-smftools-0.2.3.dist-info/entry_points.txt,sha256=q4hg4w-mKkI2leekM_-YZc5XRJzp96Mh1FcU3hac82g,52
-smftools-0.2.3.dist-info/licenses/LICENSE,sha256=F8LwmL6vMPddaCt1z1S83Kh_OZv50alTlY7BvVx1RXw,1066
-smftools-0.2.3.dist-info/RECORD,,
+smftools-0.2.4.dist-info/METADATA,sha256=BVgWPtWTeDoNF6d1IOpvXyV0IE4fI5X_fLIs4nmVvJ4,9138
+smftools-0.2.4.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
+smftools-0.2.4.dist-info/entry_points.txt,sha256=q4hg4w-mKkI2leekM_-YZc5XRJzp96Mh1FcU3hac82g,52
+smftools-0.2.4.dist-info/licenses/LICENSE,sha256=F8LwmL6vMPddaCt1z1S83Kh_OZv50alTlY7BvVx1RXw,1066
+smftools-0.2.4.dist-info/RECORD,,