smftools 0.3.2__tar.gz → 0.4.5__tar.gz

smftools-0.4.5/.claude/scheduled_tasks.lock

@@ -0,0 +1 @@
+{"sessionId":"15d14348-db85-4d8f-9fbc-7c6144caa7f9","pid":71788,"procStart":"Sun Jul 19 22:49:56 2026","acquiredAt":1784563851455}

smftools-0.4.5/.claude/settings.local.json

@@ -0,0 +1,8 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(python:*)",
+      "Bash(tail:*)"
+    ]
+  }
+}

smftools-0.4.5/.devcontainer/Dockerfile

@@ -0,0 +1,38 @@
+FROM python:3.12-slim-bookworm
+
+# System deps needed by pysam, pybedtools, htslib, etc.
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    build-essential \
+    git \
+    zlib1g-dev \
+    libbz2-dev \
+    liblzma-dev \
+    libcurl4-openssl-dev \
+    libssl-dev \
+    libdeflate-dev \
+    samtools \
+    bedtools \
+    && rm -rf /var/lib/apt/lists/*
+
+# Create non-root user
+ARG USERNAME=vscode
+ARG USER_UID=1000
+ARG USER_GID=$USER_UID
+RUN groupadd --gid $USER_GID $USERNAME \
+    && useradd --uid $USER_UID --gid $USER_GID -m $USERNAME
+
+# Set up a venv so the editable install is isolated
+RUN python -m venv /opt/venv \
+    && /opt/venv/bin/pip install --upgrade pip
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Pre-install heavy deps so rebuilds are faster
+COPY pyproject.toml README.md LICENSE /tmp/smftools/
+RUN pip install --no-cache-dir \
+    "numpy>=1.22.0,<2" \
+    "scipy>=1.7.3" \
+    "torch>=1.9.0" \
+    "anndata>=0.10.0" \
+    "pysam>=0.19.1"
+
+USER $USERNAME

smftools-0.4.5/.devcontainer/devcontainer.json

@@ -0,0 +1,32 @@
+{
+  "name": "smftools",
+  "build": {
+    "dockerfile": "Dockerfile",
+    "context": ".."
+  },
+  "customizations": {
+    "vscode": {
+      "settings": {
+        "python.defaultInterpreterPath": "/opt/venv/bin/python",
+        "python.testing.pytestEnabled": true,
+        "python.testing.pytestArgs": ["tests"],
+        "editor.formatOnSave": true,
+        "[python]": {
+          "editor.defaultFormatter": "charliermarsh.ruff",
+          "editor.codeActionsOnSave": {
+            "source.fixAll": "explicit",
+            "source.organizeImports": "explicit"
+          }
+        }
+      },
+      "extensions": [
+        "ms-python.python",
+        "ms-python.vscode-pylance",
+        "charliermarsh.ruff",
+        "ms-toolsai.jupyter"
+      ]
+    }
+  },
+  "postCreateCommand": "pip install -e '.[all,dev]'",
+  "remoteUser": "vscode"
+}

{smftools-0.3.2 → smftools-0.4.5}/.github/workflows/ci.yml

@@ -26,7 +26,7 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          python -m pip install "ruff==0.8.6"
+          python -m pip install "ruff==0.15.0"
       - name: Check formatting with ruff
         run: ruff format --check .
 
@@ -45,7 +45,7 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          python -m pip install "ruff==0.8.6"
+          python -m pip install "ruff==0.15.0"
       - name: Lint with ruff
         run: ruff check --output-format=github .
 

{smftools-0.3.2 → smftools-0.4.5}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: smftools
-Version: 0.3.2
+Version: 0.4.5
 Summary: Single Molecule Footprinting Analysis in Python.
 Project-URL: Source, https://github.com/jkmckenna/smftools
 Project-URL: Documentation, https://smftools.readthedocs.io/
@@ -47,13 +47,15 @@ Requires-Python: >=3.10
 Requires-Dist: anndata>=0.10.0
 Requires-Dist: biopython>=1.79
 Requires-Dist: click
+Requires-Dist: dask[array]
 Requires-Dist: numpy<2,>=1.22.0
 Requires-Dist: pandas>=1.4.2
+Requires-Dist: psutil>=5.9.0
 Requires-Dist: scipy>=1.7.3
 Requires-Dist: tqdm
 Provides-Extra: all
 Requires-Dist: captum; extra == 'all'
-Requires-Dist: fastcluster; extra == 'all'
+Requires-Dist: edlib; extra == 'all'
 Requires-Dist: hydra-core; extra == 'all'
 Requires-Dist: igraph; extra == 'all'
 Requires-Dist: leidenalg; extra == 'all'
@@ -63,8 +65,6 @@ Requires-Dist: multiqc; extra == 'all'
 Requires-Dist: networkx>=3.2; extra == 'all'
 Requires-Dist: omegaconf; extra == 'all'
 Requires-Dist: pod5>=0.1.21; extra == 'all'
-Requires-Dist: pybedtools>=0.12.0; extra == 'all'
-Requires-Dist: pybigwig>=0.3.24; extra == 'all'
 Requires-Dist: pynndescent>=0.5.10; extra == 'all'
 Requires-Dist: pysam>=0.19.1; extra == 'all'
 Requires-Dist: scikit-learn>=1.0.2; extra == 'all'
@@ -75,15 +75,40 @@ Requires-Dist: torch>=1.9.0; extra == 'all'
 Requires-Dist: umap-learn>=0.5.5; extra == 'all'
 Requires-Dist: upsetplot; extra == 'all'
 Requires-Dist: wandb; extra == 'all'
+Requires-Dist: xgboost>=1.7; extra == 'all'
+Provides-Extra: all-2
+Requires-Dist: captum; extra == 'all-2'
+Requires-Dist: edlib; extra == 'all-2'
+Requires-Dist: hydra-core; extra == 'all-2'
+Requires-Dist: igraph; extra == 'all-2'
+Requires-Dist: leidenalg; extra == 'all-2'
+Requires-Dist: lightning; extra == 'all-2'
+Requires-Dist: matplotlib>=3.6.2; extra == 'all-2'
+Requires-Dist: multiqc; extra == 'all-2'
+Requires-Dist: networkx>=3.2; extra == 'all-2'
+Requires-Dist: omegaconf; extra == 'all-2'
+Requires-Dist: pod5>=0.1.21; extra == 'all-2'
+Requires-Dist: pybedtools>=0.12.0; extra == 'all-2'
+Requires-Dist: pybigwig>=0.3.24; extra == 'all-2'
+Requires-Dist: pynndescent>=0.5.10; extra == 'all-2'
+Requires-Dist: pysam>=0.19.1; extra == 'all-2'
+Requires-Dist: scikit-learn>=1.0.2; extra == 'all-2'
+Requires-Dist: seaborn>=0.11; extra == 'all-2'
+Requires-Dist: shap; extra == 'all-2'
+Requires-Dist: tensorly; extra == 'all-2'
+Requires-Dist: torch>=1.9.0; extra == 'all-2'
+Requires-Dist: umap-learn>=0.5.5; extra == 'all-2'
+Requires-Dist: upsetplot; extra == 'all-2'
+Requires-Dist: wandb; extra == 'all-2'
+Requires-Dist: xgboost>=1.7; extra == 'all-2'
 Provides-Extra: cluster
-Requires-Dist: fastcluster; extra == 'cluster'
 Requires-Dist: igraph; extra == 'cluster'
 Requires-Dist: leidenalg; extra == 'cluster'
 Provides-Extra: dev
 Requires-Dist: pre-commit; extra == 'dev'
 Requires-Dist: pytest; extra == 'dev'
 Requires-Dist: pytest-cov; extra == 'dev'
-Requires-Dist: ruff; extra == 'dev'
+Requires-Dist: ruff==0.15.0; extra == 'dev'
 Provides-Extra: docs
 Requires-Dist: ipython>=7.20; extra == 'docs'
 Requires-Dist: matplotlib!=3.6.1; extra == 'docs'
@@ -108,6 +133,7 @@ Provides-Extra: ml-base
 Requires-Dist: scikit-learn>=1.0.2; extra == 'ml-base'
 Requires-Dist: tensorly; extra == 'ml-base'
 Requires-Dist: torch>=1.9.0; extra == 'ml-base'
+Requires-Dist: xgboost>=1.7; extra == 'ml-base'
 Provides-Extra: ml-extended
 Requires-Dist: captum; extra == 'ml-extended'
 Requires-Dist: hydra-core; extra == 'ml-extended'
@@ -133,6 +159,10 @@ Requires-Dist: torch>=1.9.0; extra == 'torch'
 Provides-Extra: umap
 Requires-Dist: pynndescent>=0.5.10; extra == 'umap'
 Requires-Dist: umap-learn>=0.5.5; extra == 'umap'
+Provides-Extra: umi
+Requires-Dist: edlib; extra == 'umi'
+Provides-Extra: xgboost
+Requires-Dist: xgboost>=1.7; extra == 'xgboost'
 Description-Content-Type: text/markdown
 
 [![PyPI](https://img.shields.io/pypi/v/smftools.svg)](https://pypi.org/project/smftools)

smftools-0.4.5/docs/source/api/analysis.md

@@ -0,0 +1,65 @@
+## Analysis: `an`
+
+Downstream statistical analysis library. Pure functions with no project-specific knowledge,
+organised into four subpackages by role:
+
+| Subpackage | Input | Output |
+|---|---|---|
+| `compute` | numpy arrays / DataFrames | arrays / dicts (no I/O) |
+| `plot` | results + `output_path` | figure written to disk |
+| `filters` | obs or var DataFrame | boolean `np.ndarray` mask |
+| `config` | none (static) | configuration dicts |
+
+### Compute
+
+Pure statistical compute functions. No AnnData dependency (except `ep_classification`).
+
+```{eval-rst}
+.. autosummary::
+   :toctree: generated/analysis_compute
+
+   smftools.analysis.compute.autocorrelation
+   smftools.analysis.compute.pearson
+   smftools.analysis.compute.hmm_features
+   smftools.analysis.compute.ep_classification
+   smftools.analysis.compute.ls_periodicity
+   smftools.analysis.compute.dimensionality_reduction
+   smftools.analysis.compute.read_cache
+```
+
+### Plot
+
+Figure rendering. Accepts results and an explicit `output_path`; writes a figure to disk.
+
+```{eval-rst}
+.. autosummary::
+   :toctree: generated/analysis_plot
+
+   smftools.analysis.plot.heatmaps
+   smftools.analysis.plot.histograms
+   smftools.analysis.plot.autocorr
+   smftools.analysis.plot.locus
+```
+
+### Filters
+
+Boolean mask builders for obs-level and var-level (position) selection.
+
+```{eval-rst}
+.. autosummary::
+   :toctree: generated/analysis_filters
+
+   smftools.analysis.filters.obs_filters
+   smftools.analysis.filters.position_filters
+```
+
+### Config
+
+Static configuration objects; no runtime inputs.
+
+```{eval-rst}
+.. autosummary::
+   :toctree: generated/analysis_config
+
+   smftools.analysis.config.hmm_histogram
+```

{smftools-0.3.2 → smftools-0.4.5}/docs/source/api/index.md

@@ -12,6 +12,7 @@ import smftools as smf
 informatics
 preprocessing
 tools
+analysis
 datasets
 ```
 

{smftools-0.3.2 → smftools-0.4.5}/docs/source/conf.py

@@ -82,7 +82,6 @@ autodoc_mock_imports = [
     "anndata",
     "Bio",
     "captum",
-    "fastcluster",
     "hydra",
     "igraph",
     "leidenalg",

{smftools-0.3.2 → smftools-0.4.5}/docs/source/tutorials/cli_usage.md

@@ -29,7 +29,7 @@ The load command builds the raw AnnData object from your raw sequencing data. It
 - adata.X contains binarized modification data (conversion/deaminase), or modification probabilitiesc (native).
 - Adds basic read-level QC annotations (Read start, end, length, mean quality).
 - Adds layers encoding read DNA sequences, base quality scores, base mismatches.
-- Maintains BAM Tags/Flags in adata.obs.
+- Maintains BAM tags/flags in adata.obs (UMI and barcode annotations loaded from Parquet sidecars).
 - Writes the raw AnnData to the canonical output path and runs MultiQC.
 - Optionally deletes intermediate BAMs, H5ADs, and TSVs.
 
@@ -52,6 +52,26 @@ The preprocess command performs QC, binarization, filtering, and duplicate detec
 - Visualizes read span masks and base quality scores with clustermaps.
 - Writes preprocessed (duplicates flagged, but kept) and preprocessed/deduplicated AnnData outputs.
 
+General AnnData structures added by `preprocess`:
+
+- `obs`
+- sample-sheet metadata columns (when `sample_sheet_path` is provided) mapped by `sample_sheet_mapping_column`.
+- read-level QC/ratio/modification summary fields used for filtering and plotting (for example read length/quality/mapping and fraction-modified metrics).
+- optional UMI preprocessing fields when `use_umi=True`, including validity and clustering annotations (for example `U1_valid`, `U2_valid`, `U1_cluster`, `U2_cluster`, `RX_cluster`).
+- optional UMI bipartite graph annotations and dominance metrics (for example edge-count/dominant-pair style fields) plus group-level stats in `uns`.
+- duplicate-detection fields for non-direct modalities (for example merged cluster IDs/sizes and duplicate flags).
+- `var`
+- per-reference position masks and site-context columns used downstream (for example `position_in_<reference>` and `<reference>_<site_type>_site`).
+- optional reindex columns from `reindex_references_adata` (suffix controlled by `reindexed_var_suffix`).
+- `layers`
+- binarized signal layer for direct modality (name controlled by `output_binary_layer_name`).
+- NaN-cleaning strategy layers (for example `fill_nans_closest`, `nan0_0minus1`, `nan1_12`, `nan_minus_1`, `nan_half`).
+- base-context/site-type derived binary layers used for downstream analyses and duplicate detection.
+- `obsm`
+- base-context level arrays written by context appending steps for downstream plotting/analysis.
+- `uns`
+- preprocessing stage metadata/flags plus auxiliary analysis outputs (for example UMI bipartite summaries and complexity-analysis summaries/fit outputs).
+
 ### `smftools variant`
 
 The variant command focuses on DNA sequence variation analyses. It:
@@ -61,6 +81,24 @@ The variant command focuses on DNA sequence variation analyses. It:
 - Generates z-scores for variant occurance given read level Q-scores and assuming uniform Palt transitions.
 - Visualizes read DNA sequence encodings and mismatch encodings.
 
+General AnnData structures added by `variant`:
+
+- `layers`
+- `"{seq1_col}__{seq2_col}_variant_call"`: per-position variant call state (`1=seq1`, `2=seq2`, `0=unknown/no-coverage`, `-1=non-informative/non-mismatch`).
+- `"{seq1_col}__{seq2_col}_variant_segments"`: segmented track per read span (`0=outside span`, `1=seq1 segment`, `2=seq2 segment`, `3=transition zone`).
+- `var`
+- `"{prefix}_seq1_acceptable_bases"` and `"{prefix}_seq2_acceptable_bases"`: accepted base sets used for variant matching at informative sites.
+- `"{prefix}_informative_site"`: boolean mask of informative mismatch positions.
+- `obs`
+- `"{prefix}_breakpoint_count"` and `"{prefix}_is_chimeric"`: per-read breakpoint summary.
+- `"{prefix}_variant_breakpoints"` and `variant_breakpoints`: list of inferred breakpoint positions per read.
+- `chimeric_variant_sites` and `chimeric_variant_sites_type`: mismatch-segment chimera flags and categorical type labels.
+- `"{prefix}_variant_segment_cigar"` and `variant_segment_cigar`: run-length string using `S` (self) and `X` (other).
+- `"{prefix}_variant_self_base_count"` / `variant_self_base_count`: count of self-classified bases per read span.
+- `"{prefix}_variant_other_base_count"` / `variant_other_base_count`: count of other-classified bases per read span.
+- `uns`
+- workflow completion flags (e.g., `append_variant_call_layer_performed`, `append_variant_segment_layer_performed`) and prior mismatch/substitution metadata used by variant calling.
+
 ### `smftools chimeric`
 
 The chimeric command is meant to find putative PCR chimeras. It:
@@ -69,7 +107,22 @@ The chimeric command is meant to find putative PCR chimeras. It:
 - Performs sliding window nearest neighbor hamming distance analysis per read.
 - Visualizes the windowed nearest neighbor hamming distances per read.
 - Assembles maximum spanning intervals of 0-hamming distance neighbors per read within the reference/sample.
-- In progress.
+
+General AnnData structures added by `chimeric`:
+
+- `obsm`
+- `cfg.rolling_nn_obsm_key`: per-read rolling nearest-neighbor hamming distance tracks.
+- `layers`
+- `zero_hamming_distance_spans`: within-sample/reference top span mask derived from zero-distance partners.
+- `cross_sample_zero_hamming_distance_spans`: top span mask from cross-sample pooling.
+- `delta_zero_hamming_distance_spans`: clipped difference (`within - cross`) used for delta-based chimera evidence.
+- `obs`
+- `chimeric_by_mod_hamming_distance`: boolean flag based on longest positive delta span threshold.
+- per-read top-segment tuple lists under keys like `"{rolling_nn_obsm_key}__top_segments"` (when top-segment extraction is enabled).
+- `uns`
+- rolling-distance and zero-pair metadata keyed by `rolling_nn_obsm_key`, including maps such as:
+- `"{rolling_nn_obsm_key}_zero_pairs_map"` and `"{rolling_nn_obsm_key}_reference_map"`.
+- optional stored segment records (`...__zero_hamming_segments`) and plotting metadata (`..._starts`, `..._window`, `..._step`, etc.), depending on cleanup settings in config.
 
 ### `smftools spatial`
 
@@ -133,4 +186,4 @@ smftools batch spatial /path/to/configs.tsv --column my_config --sep $'\t'
 ```
 
 Each path is validated; missing configs are skipped with a message, while valid configs run the
-requested task in sequence.
+requested task in sequence.

smftools-0.4.5/docs/source/tutorials/experiment_config.md

@@ -0,0 +1,99 @@
+# Experiment configuration CSV
+
+smftools uses an experiment configuration CSV to define paths, modality settings, and workflow
+options. You can start from the repository template (`experiment_config.csv`) and fill in your
+experiment-specific values. The configuration CSV can override any parameter within the default.yaml
+and modality specific config .yamls found within the config subpackage of smftools.
+
+## CSV format
+
+The configuration CSV is a table with the following columns:
+
+| Column | Description |
+| --- | --- |
+| `variable` | Configuration key name (used by smftools). |
+| `value` | Your value for this key. |
+| `help` | Short description of the key. |
+| `options` | Expected values (when applicable). |
+| `type` | Expected value type (`str`, `int`, `float`, `list`). |
+
+A shortened example looks like:
+
+```text
+variable,value,help,options,type
+smf_modality,conversion,Modality of SMF. Can either be conversion or direct.,"conversion, direct",str
+input_data_path,/path_to_POD5_directory,Path to directory/file containing input sequencing data,,str
+fasta,/path_to_fasta.fasta,Path to initial FASTA file,,str
+output_directory,/outputs,Directory to act as root for all analysis outputs,,str
+experiment_name,,An experiment name for the final h5ad file,,str
+```
+
+## Common fields
+
+Below are some of the most commonly edited fields and how they affect the CLI workflows:
+
+- `smf_modality`: Defines whether the data is `conversion`, `direct` or `deaminase`, which determines
+  preprocessing and HMM feature handling.
+- `input_data_path`: Location of raw input data (fast5/pod5/fastq/bam).
+- `fasta`: Reference FASTA for alignment and positional context.
+- `fasta_regions_of_interest`: Optional BED file to subset the FASTA.
+- `output_directory`: Root output folder for all generated AnnData files and plots.
+- `experiment_name`: Base name used for output AnnData files.
+- `model_dir` / `model`: Dorado basecalling model configuration (nanopore runs).
+- `demux_backend`: Demultiplexing backend (`dorado` or `smftools`).
+- `barcode_kit`: Barcode kit name. Required for `dorado`; for `smftools`, use either a known alias or
+  `custom` plus `custom_barcode_yaml`.
+- `custom_barcode_yaml`: Barcode reference YAML path used when `demux_backend=smftools` and
+  `barcode_kit=custom`.
+- `use_umi` / `umi_yaml`: Optional UMI extraction controls. `umi_yaml` can define flanking-aware UMI
+  extraction.
+- `mapping_threshold`: Minimum mapping proportion per reference required for downstream steps.
+- `mod_list`: Modification calls to use for direct-modality workflows.
+- `conversion_types`: Target modification types for conversion workflows.
+
+## Tips
+
+- Keep paths absolute whenever possible to avoid ambiguity.
+- Lists are written in bracketed form, e.g. `[5mC]` or `[5mC_5hmC]`.
+- If you update the CSV, re-run the CLI command pointing at the updated file.
+
+## Read annotations
+
+smftools annotates reads during `load_adata` and stores the results in `adata.obs`. Standard BAM
+tags (e.g. `NM`, `MD`, `MM`, `ML`) are read directly from BAM files. UMI and barcode annotations
+are computed in parallel and written to Parquet sidecar files alongside the aligned BAM, then loaded
+into `adata.obs` from those sidecars. The aligned BAM itself is not modified.
+
+**UMI annotations** (written to `.umi_tags.parquet`)
+
+- `U1`: Orientation-corrected UMI for the *left* reference end of the mapped fragment (forward reads: US, reverse reads: UE).
+- `U2`: Orientation-corrected UMI for the *right* reference end of the mapped fragment (forward reads: UE, reverse reads: US).
+- `US`: Positional UMI from read start (delimited `UMI_seq;slot;flank_seq`).
+- `UE`: Positional UMI from read end (delimited `UMI_seq;slot;flank_seq`).
+- `RX`: Combined UMI string (`U1-U2`, or `U1`/`U2` if only one is present).
+- `FC`: Flank context of the U1/U2 pair (e.g. `top-bottom`).
+
+When `threads` is set, UMI extraction is parallelized across multiple CPU cores.
+
+**Barcode annotations (smftools demux backend)** (written to `.barcode_tags.parquet`)
+
+- `BC`: Assigned barcode name, or `unclassified`.
+- `BM`: Match type (`both`, `read_start_only`, `read_end_only`, `mismatch`, `unclassified`).
+- `B1`: Edit distance for the read-start barcode match.
+- `B2`: Edit distance for the read-end barcode match.
+- `B3`: Extracted barcode sequence from the read start (forward orientation).
+- `B4`: Extracted barcode sequence from the read end (reverse-complemented to forward orientation).
+- `B5`: Barcode name matched at the read start (corresponds to `B1`/`B3`).
+- `B6`: Barcode name matched at the read end (corresponds to `B2`/`B4`).
+
+When `threads` is set, barcode extraction is parallelized across multiple CPU cores.
+Demultiplexing (splitting reads into per-barcode BAMs) uses the sidecar `BC` assignments.
+Only primary alignments are included in split BAMs and sidecar files.
+
+**Barcode annotations (dorado demux backend)**
+
+- `BC`: Assigned barcode name (read from BAM tag).
+- `bi`: Dorado barcode info array (if present; expanded into columns during load).
+
+Notes:
+- `BE`/`BF` are not used by smftools.

{smftools-0.3.2 → smftools-0.4.5}/experiment_config.csv

@@ -1,14 +1,14 @@
-variable,value,help,options,type
-smf_modality,conversion,Modality of SMF. Can either be conversion or direct.,"conversion, direct",str 
-input_data_path,/path_to_POD5_directory,Path to directory/file containing input sequencing data,,str 
-fasta,/path_to_fasta.fasta,Path to initial FASTA file,,str 
-fasta_regions_of_interest,/path_to_bed.bed,Path to a bed file to subsample the fasta on.,,str 
-output_directory,/outputs,Directory to act as root for all analysis outputs,,str 
-experiment_name,,An experiment name for the final h5ad file,,str 
-model_dir,/path_to_dorado_model_dir,Path,,str 
-model,None,The dorado basecalling model to use,,str 
-barcode_kit,SQK-NBD114-24,The barcoding kit used for the experiment,,str 
-mapping_threshold,0.05,Minimum proportion of reads mapping to a reference to further use that reference (Ranges from 0-1 as a proportion of mapped reads),,float
-mod_list,[5mC_5hmC],Modified base names for Dorado,"""6mA"", ""5mC_5hmC""",list
-batch_size,4,number of samples to analyze at a time,,int
+variable,value,help,options,type
+smf_modality,conversion,Modality of SMF. Can either be conversion or direct.,"conversion, direct",str 
+input_data_path,/path_to_POD5_directory,Path to directory/file containing input sequencing data,,str 
+fasta,/path_to_fasta.fasta,Path to initial FASTA file,,str 
+fasta_regions_of_interest,/path_to_bed.bed,Path to a bed file to subsample the fasta on.,,str 
+output_directory,/outputs,Directory to act as root for all analysis outputs,,str 
+experiment_name,,An experiment name for the final h5ad file,,str 
+model_dir,/path_to_dorado_model_dir,Path,,str 
+model,None,The dorado basecalling model to use,,str 
+barcode_kit,SQK-NBD114-24,The barcoding kit used for the experiment,,str 
+mapping_threshold,0.05,Minimum proportion of reads mapping to a reference to further use that reference (Ranges from 0-1 as a proportion of mapped reads),,float
+mod_list,[5mC_5hmC],Modified base names for Dorado,"""6mA"", ""5mC_5hmC""",list
+batch_size,4,number of samples to analyze at a time,,int
 conversion_types,[5mC],Types of modification types to use in conversion SMF,"5mC', '6mA'",list

{smftools-0.3.2 → smftools-0.4.5}/notebooks/Kissiov_and_McKenna_2025_sample_sheet.csv

@@ -1,11 +1,11 @@
-Sample,Sample_names,MTase,Time (sec),Group
-0,,Hia5,0,0
-1,,Hia5,15,0
-2,,Hia5,30,0
-3,,Hia5,120,0
-4,,Hia5,300,0
-5,,Hia5,0,1
-6,,Hia5,15,1
-7,,Hia5,30,1
-8,,Hia5,120,1
+Sample,Sample_names,MTase,Time (sec),Group
+0,,Hia5,0,0
+1,,Hia5,15,0
+2,,Hia5,30,0
+3,,Hia5,120,0
+4,,Hia5,300,0
+5,,Hia5,0,1
+6,,Hia5,15,1
+7,,Hia5,30,1
+8,,Hia5,120,1
 9,,Hia5,300,1

{smftools-0.3.2 → smftools-0.4.5}/pyproject.toml

@@ -43,8 +43,10 @@ dependencies = [
     "anndata>=0.10.0",
     "biopython>=1.79",
     "click",
+    "dask[array]",
     "numpy>=1.22.0,<2",
     "pandas>=1.4.2",
+    "psutil>=5.9.0",
     "scipy>=1.7.3",
     "tqdm",
 ]
@@ -60,9 +62,9 @@ smftools = "smftools.cli_entry:cli"
 [project.optional-dependencies]
 
 dev = [
-    "ruff", 
-    "pre-commit", 
-    "pytest", 
+    "ruff==0.15.0",
+    "pre-commit",
+    "pytest",
     "pytest-cov"]
 
 docs = [
@@ -85,7 +87,6 @@ docs = [
 ]
 
 cluster = [
-    "fastcluster",
     "igraph",
     "leidenalg",
 ]
@@ -120,6 +121,11 @@ ml-base = [
     "scikit-learn>=1.0.2",
     "tensorly",
     "torch>=1.9.0",
+    "xgboost>=1.7",
+]
+
+xgboost = [
+    "xgboost>=1.7",
 ]
 
 ml-extended = [
@@ -135,6 +141,10 @@ qc = [
     "multiqc",
 ]
 
+umi = [
+    "edlib",
+]
+
 umap = [
     "pynndescent>=0.5.10",
     "umap-learn>=0.5.5",
@@ -146,7 +156,48 @@ torch = [
 
 all = [
     # cluster
-    "fastcluster",
+    "igraph",
+    "leidenalg",
+
+    # informatics
+    "pysam>=0.19.1",
+
+    # misc
+    "networkx>=3.2",
+    "upsetplot",
+
+    # ont
+    "pod5>=0.1.21",
+
+    # plotting
+    "matplotlib>=3.6.2",
+    "seaborn>=0.11",
+
+    # ml
+    "captum",
+    "hydra-core",
+    "lightning",
+    "omegaconf",
+    "scikit-learn>=1.0.2",
+    "shap",
+    "tensorly",
+    "torch>=1.9.0",
+    "wandb",
+    "xgboost>=1.7",
+
+    # umap
+    "pynndescent>=0.5.10",
+    "umap-learn>=0.5.5",
+
+    # qc
+    "multiqc",
+
+    # umi
+    "edlib",
+]
+
+all_2 = [
+    # cluster
     "igraph",
     "leidenalg",
 
@@ -176,6 +227,7 @@ all = [
     "tensorly",
     "torch>=1.9.0",
     "wandb",
+    "xgboost>=1.7",
 
     # umap
     "pynndescent>=0.5.10",
@@ -183,6 +235,9 @@ all = [
 
     # qc
     "multiqc",
+
+    # umi
+    "edlib",
 ]
 
 [tool.hatch.build]

{smftools-0.3.2 → smftools-0.4.5}/requirements.txt

@@ -10,7 +10,6 @@ seaborn>=0.11
 tqdm
 
 # Graph / clustering
-fastcluster
 igraph
 leidenalg
 networkx>=3.2
@@ -33,6 +32,7 @@ shap
 tensorly
 torch>=1.9.0
 wandb
+xgboost>=1.7
 
 # QC / viz
 multiqc