multibench-sc 0.2.1__tar.gz

multibench_sc-0.2.1/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

multibench_sc-0.2.1/PKG-INFO

@@ -0,0 +1,83 @@
+Metadata-Version: 2.4
+Name: multibench-sc
+Version: 0.2.1
+Summary: Clean Python API wrapping the scMultiBench benchmark: run methods, evaluate, and plot.
+License: Apache-2.0
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas<3,>=2.0
+Requires-Dist: numpy>=1.24
+Requires-Dist: matplotlib>=3.7
+Requires-Dist: h5py>=3.8
+Requires-Dist: anndata>=0.9
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: scanpy>=1.9
+Requires-Dist: scib>=1.1
+Provides-Extra: dev
+Requires-Dist: pytest>=7.4; extra == "dev"
+Provides-Extra: eval
+Provides-Extra: loom
+Requires-Dist: loompy>=3.0; extra == "loom"
+Dynamic: license-file
+
+# scMultiBench
+
+[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/DSichang/scMultiBench/blob/main/notebooks/colab_quickstart.ipynb)
+[![Docs](https://img.shields.io/badge/docs-dsichang.github.io%2FscMultiBench-blue)](https://dsichang.github.io/scMultiBench/)
+[![PyPI](https://img.shields.io/pypi/v/multibench-sc)](https://pypi.org/project/multibench-sc/)
+
+A systematic benchmark of single-cell multimodal integration, with `multibench` -
+a typed Python API that runs 40 integration methods across four scenarios
+(vertical, diagonal, mosaic, cross), scores them with scIB metrics, and draws
+the paper's figures.
+
+**Documentation and tutorials:** <https://dsichang.github.io/scMultiBench/>
+
+## Quick start
+
+```bash
+pip install multibench-sc          # the API (import name: multibench)
+```
+
+For the tutorials (they also use the stored benchmark tables shipped in this
+repository), clone instead: `git clone https://github.com/DSichang/scMultiBench.git
+&& cd scMultiBench && pip install -e .`
+
+```python
+import multibench as mtb
+
+mtb.list_methods()                 # the 40-method registry
+mtb.method_info("Matilda")         # everything known about one method
+mtb.scan("D11", "vertical")        # what can run on a dataset, and why not
+res = mtb.run_all("D11", "vertical", out_dir="out/")   # run + score
+res.plot()                         # the paper-style bubble panel
+```
+
+Running methods needs their conda environments (Linux). The package itself is
+~2 MB - install only the environments you need:
+
+```bash
+multibench env doctor                              # what exists / is missing
+multibench env install --methods Matilda --run     # one method (2-14 GB)
+multibench env install --category vertical --run   # one category (45-101 GB)
+```
+
+The benchmark datasets are downloaded separately - see
+[Get the data](https://dsichang.github.io/scMultiBench/installation/#get-the-data).
+
+## Try it without installing anything
+
+The [Colab quickstart](https://colab.research.google.com/github/DSichang/scMultiBench/blob/main/notebooks/colab_quickstart.ipynb)
+installs the API, explores the registry, and reproduces the benchmark figures
+from the shipped result tables - entirely in the browser. The full published
+rankings are browsable in the
+[interactive explorer](https://shiny.maths.usyd.edu.au/scMultiBench/).
+
+## Citation
+
+Liu, Ding et al. Benchmarking single-cell multimodal data integrations.
+*Nature Methods* 22, 2449-2460 (2025).
+Every method you run is third-party software with its own paper - please cite
+it alongside the benchmark; `mtb.method_info(name)` points to each method's
+upstream repository and reference.

multibench_sc-0.2.1/README.md

@@ -0,0 +1,60 @@
+# scMultiBench
+
+[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/DSichang/scMultiBench/blob/main/notebooks/colab_quickstart.ipynb)
+[![Docs](https://img.shields.io/badge/docs-dsichang.github.io%2FscMultiBench-blue)](https://dsichang.github.io/scMultiBench/)
+[![PyPI](https://img.shields.io/pypi/v/multibench-sc)](https://pypi.org/project/multibench-sc/)
+
+A systematic benchmark of single-cell multimodal integration, with `multibench` -
+a typed Python API that runs 40 integration methods across four scenarios
+(vertical, diagonal, mosaic, cross), scores them with scIB metrics, and draws
+the paper's figures.
+
+**Documentation and tutorials:** <https://dsichang.github.io/scMultiBench/>
+
+## Quick start
+
+```bash
+pip install multibench-sc          # the API (import name: multibench)
+```
+
+For the tutorials (they also use the stored benchmark tables shipped in this
+repository), clone instead: `git clone https://github.com/DSichang/scMultiBench.git
+&& cd scMultiBench && pip install -e .`
+
+```python
+import multibench as mtb
+
+mtb.list_methods()                 # the 40-method registry
+mtb.method_info("Matilda")         # everything known about one method
+mtb.scan("D11", "vertical")        # what can run on a dataset, and why not
+res = mtb.run_all("D11", "vertical", out_dir="out/")   # run + score
+res.plot()                         # the paper-style bubble panel
+```
+
+Running methods needs their conda environments (Linux). The package itself is
+~2 MB - install only the environments you need:
+
+```bash
+multibench env doctor                              # what exists / is missing
+multibench env install --methods Matilda --run     # one method (2-14 GB)
+multibench env install --category vertical --run   # one category (45-101 GB)
+```
+
+The benchmark datasets are downloaded separately - see
+[Get the data](https://dsichang.github.io/scMultiBench/installation/#get-the-data).
+
+## Try it without installing anything
+
+The [Colab quickstart](https://colab.research.google.com/github/DSichang/scMultiBench/blob/main/notebooks/colab_quickstart.ipynb)
+installs the API, explores the registry, and reproduces the benchmark figures
+from the shipped result tables - entirely in the browser. The full published
+rankings are browsable in the
+[interactive explorer](https://shiny.maths.usyd.edu.au/scMultiBench/).
+
+## Citation
+
+Liu, Ding et al. Benchmarking single-cell multimodal data integrations.
+*Nature Methods* 22, 2449-2460 (2025).
+Every method you run is third-party software with its own paper - please cite
+it alongside the benchmark; `mtb.method_info(name)` points to each method's
+upstream repository and reference.

multibench_sc-0.2.1/multibench/__init__.py

@@ -0,0 +1,46 @@
+"""multibench - run and compare single-cell multimodal integration methods.
+
+Quickstart
+----------
+::
+
+    import multibench as mtb
+
+    mtb.list_categories()            # the four integration scenarios
+    print(mtb.describe_layout())     # how to lay out YOUR dataset
+
+    mtb.scan("MYDATA")               # which methods can run on it (and why not the rest)
+
+    res = mtb.run_all("MYDATA", "vertical", out_dir="out/")
+    print(res.summary)               # per-method status + metrics
+    print(res.failures)              # what did not finish, and why
+    res.plot().savefig("compare.png")
+
+The four ``category`` values are ``vertical``, ``diagonal``, ``mosaic`` and
+``cross`` - see :func:`list_categories` for what each one means, and
+:func:`describe_layout` for the filenames each expects.
+"""
+
+from . import config
+from . import plot
+from . import eval
+from .data import catalog
+from .data.results import load_results, available_datasets
+from .eval.pipeline import evaluate, to_long
+from .engine.runner import run
+from .engine.registry import list_methods, list_tasks
+from .engine.resolve import inputs_for, labels_for
+from .engine import ingest as io
+from .engine import envs as env
+from .discover import find_methods, method_info, params_for
+from .workflow import (scan, run_all, BatchResult, list_categories,
+                       describe_layout, load_batch, runtime_hint, sweep)
+
+__version__ = "0.2.1"
+
+__all__ = ["config", "plot", "eval", "catalog", "load_results", "available_datasets",
+           "evaluate", "to_long", "run", "list_methods", "list_tasks", "inputs_for",
+           "labels_for", "io", "env", "method_info", "find_methods", "params_for",
+           "scan", "run_all", "BatchResult",
+           "list_categories", "describe_layout", "load_batch",
+           "runtime_hint", "sweep", "__version__"]

multibench_sc-0.2.1/multibench/__main__.py

@@ -0,0 +1,7 @@
+"""Enable ``python -m multibench`` (equivalent to the ``multibench`` console script)."""
+import sys
+
+from .cli import main
+
+if __name__ == "__main__":
+    sys.exit(main())

multibench_sc-0.2.1/multibench/cli.py

@@ -0,0 +1,255 @@
+"""Thin CLI over the multibench Python API."""
+from __future__ import annotations
+
+import argparse
+import shlex
+import os
+import sys
+
+
+
+def _cmd_list(args) -> int:
+    from .engine import registry
+    for m in registry.list_methods(category=args.category, task=args.task,
+                                   runnable=args.runnable or None):
+        print(m)
+    return 0
+
+
+def _cmd_find(args) -> int:
+    needs = True if args.needs_labels else None
+    from . import discover
+    for m in discover.find_methods(category=args.category, task=args.task,
+                                   needs_labels=needs, atac=args.atac,
+                                   runnable=args.runnable or None):
+        print(m)
+    return 0
+
+
+def _cmd_plot(args) -> int:
+    if args.kind != "bubble":
+        raise SystemExit(f"unknown plot kind {args.kind!r}")
+    from . import load_results
+    from . import plot as plot_ns
+    df = load_results(category=args.category, dataset=args.dataset)
+    metrics = args.metrics.split(",") if args.metrics else None
+    plot_ns.bubble(df, metrics=metrics, aggregate=args.aggregate, save=args.out)
+    print(f"wrote {args.out}")
+    return 0
+
+
+def _parse_inputs(pairs) -> dict:
+    out = {}
+    for p in pairs or []:
+        if "=" not in p:
+            raise SystemExit(f"--input must be role=path, got {p!r}")
+        role, path = p.split("=", 1)
+        if role.strip() == "" or path == "":
+            raise SystemExit(f"--input must be role=path with non-empty parts, got {p!r}")
+        out[role] = path
+    return out
+
+
+def _cmd_run(args) -> int:
+    import multibench
+    res = multibench.run(method=args.method, category=args.category, task=args.task,
+                       inputs=_parse_inputs(args.input), out_dir=args.out,
+                       cmd_template=args.runner)
+    print(f"ran {args.method} -> {res.out_dir}")
+    return 0
+
+
+def _cmd_evaluate(args) -> int:
+    import multibench
+    df = multibench.evaluate(output=args.output, category=args.category, task=args.task,
+                           labels=args.labels, clustering=args.cluster,
+                           batch=getattr(args, "batch", None))
+    if args.out:
+        df.to_csv(args.out)
+        print(f"wrote {args.out}")
+    else:
+        print(df.to_string())
+    return 0
+
+
+def _cmd_env(args) -> int:
+    from .engine import envs
+    cmd = args.env_cmd
+    if cmd == "status":
+        for r in envs.status():
+            mark = "x" if r["exists"] else " "
+            tag = r["difficulty"] + ("*" if r["verified_working"] else "")
+            print(f"[{mark}] {r['method']:16} {r['group']:16} {tag}")
+        return 0
+    if cmd == "groups":
+        for name, spec in envs.groups().items():
+            if spec.get("shared"):
+                print(f"{name:16} ({len(spec['members']):2}): {', '.join(spec['members'])}")
+        return 0
+    if cmd == "plan":
+        _mlist = getattr(args, "methods", None)
+        _mlist = [m.strip() for m in _mlist.split(",")] if _mlist else None
+        for p in envs.plan(category=getattr(args, "category", None), methods=_mlist):
+            tag = "shared" if p["shared"] else "own"
+            print(f"{p['env']:16} [{tag:6}] <- {', '.join(p['methods'])}")
+        return 0
+    if cmd == "doctor":
+        _mlist = getattr(args, "methods", None)
+        _mlist = [m.strip() for m in _mlist.split(",")] if _mlist else None
+        rows = envs.doctor(category=getattr(args, "category", None), methods=_mlist)
+        for r in rows:
+            mark = "x" if r["exists"] else ("L" if r["has_lock"] else "!")
+            print(f"[{mark}] {r['env']:18} ({len(r['methods']):2}) <- {', '.join(r['methods'])}")
+        missing = [r for r in rows if not r["exists"]]
+        nolock = [r["env"] for r in missing if not r["has_lock"]]
+        print(f"# {len(rows)} envs needed, {len(missing)} missing"
+              + (f"; NO lockfile for: {', '.join(nolock)}" if nolock else ""))
+        print("# legend: [x]=installed  [L]=missing, lockfile ready (run `multibench env install --run`)  [!]=missing, no lockfile")
+        return 0
+    if cmd == "install":
+        _mlist = getattr(args, "methods", None)
+        _mlist = [m.strip() for m in _mlist.split(",")] if _mlist else None
+        if getattr(args, "packed", False) and getattr(args, "run", False):
+            for r in envs.doctor(category=getattr(args, "category", None),
+                                 methods=_mlist):
+                if not r["exists"] and envs.install_packed(r["env"]):
+                    print(f"{r['env']:18} [PACKED        ] <- {', '.join(r['methods'])}")
+        try:
+            rows = envs.create_all(category=getattr(args, "category", None),
+                                   methods=_mlist,
+                                   dry_run=not getattr(args, "run", False))
+        except RuntimeError as e:
+            print(f"error: {e}")
+            return 1
+        for r in rows:
+            state = ("have" if r["exists"]
+                     else ("BUILD" if r["has_lock"] and getattr(args, "run", False)
+                           else "build(dry-run)" if r["has_lock"] else "NO-LOCK"))
+            print(f"{r['env']:18} [{state:14}] <- {', '.join(r['methods'])}")
+        if not getattr(args, "run", False):
+            print("# dry-run — add --run to create the missing envs from their lockfiles")
+        return 0
+    if cmd == "freeze":
+        if getattr(args, "all", False):
+            for env in envs.required_envs(category=getattr(args, "category", None)):
+                try:
+                    print(f"froze {env} -> {envs.freeze(env)}")
+                except Exception as e:  # noqa: BLE001 - report per-env, keep going
+                    print(f"SKIP {env}: {str(e)[:120]}")
+        else:
+            if not getattr(args, "env", None):
+                print("error: name an env to freeze, or pass --all")
+                return 2
+            print(f"froze {args.env} -> {envs.freeze(args.env)}")
+        return 0
+    if cmd == "create-group":
+        cmds = envs.create_group(args.group, dry_run=not getattr(args, "run", False))
+        if not getattr(args, "run", False):
+            print("# dry-run — add --run to execute:")
+            for c in cmds:
+                print(shlex.join(c))
+        else:
+            print(f"created group env {args.group}")
+        return 0
+    method = getattr(args, "method", None)
+    name = getattr(args, "name", None)
+    if cmd == "recipe":
+        for c in envs.create_commands(method, env_name=name):
+            print(shlex.join(c))
+        return 0
+    if cmd == "yml":
+        y = envs.environment_yml(method, env_name=name)
+        out = getattr(args, "out", None)
+        if out:
+            with open(out, "w") as f:
+                f.write(y)
+            print(f"wrote {out}")
+        else:
+            print(y, end="")
+        return 0
+    if cmd == "create":
+        cmds = envs.create(method, env_name=name, dry_run=not getattr(args, "run", False))
+        if not getattr(args, "run", False):
+            print("# dry-run — add --run to execute:")
+            for c in cmds:
+                print(shlex.join(c))
+        else:
+            print(f"created environment for {method}")
+        return 0
+    raise SystemExit(
+        "usage: multibench env {status|groups|plan|doctor|install|freeze|recipe|yml|create|create-group}")
+
+
+def build_parser() -> argparse.ArgumentParser:
+    p = argparse.ArgumentParser(prog="multibench")
+    sub = p.add_subparsers(dest="command", required=True)
+
+    pl = sub.add_parser("list"); pl.add_argument("--category"); pl.add_argument("--task")
+    pl.add_argument("--runnable", action="store_true",
+                    help="only methods with a declared variant (usable by run)")
+    pl.set_defaults(func=_cmd_list)
+
+    pf = sub.add_parser("find")
+    pf.add_argument("--category"); pf.add_argument("--task")
+    pf.add_argument("--needs-labels", action="store_true")
+    pf.add_argument("--atac", choices=["peak", "gene_activity"],
+                    help="filter by ATAC representation the method consumes")
+    pf.add_argument("--runnable", action="store_true",
+                    help="only methods with a declared variant (usable by run)")
+    pf.set_defaults(func=_cmd_find)
+
+    pp = sub.add_parser("plot")
+    pp.add_argument("kind")
+    pp.add_argument("--category", required=True); pp.add_argument("--dataset")
+    pp.add_argument("--metrics"); pp.add_argument("--aggregate", default="dataset")
+    pp.add_argument("--out", required=True)
+    pp.set_defaults(func=_cmd_plot)
+
+    pr = sub.add_parser("run")
+    pr.add_argument("--method", required=True); pr.add_argument("--category", required=True)
+    pr.add_argument("--task", default="clustering")
+    pr.add_argument("--input", action="append", help="role=path (repeatable)")
+    pr.add_argument("--out", required=True); pr.add_argument("--runner")
+    pr.set_defaults(func=_cmd_run)
+
+    pe = sub.add_parser("evaluate")
+    pe.add_argument("--output", required=True)
+    pe.add_argument("--category", help="reserved; not used by v1 metrics")
+    pe.add_argument("--task", default="clustering")
+    pe.add_argument("--batch", help="per-cell batch labels CSV (required for --task batch/all)")
+    pe.add_argument("--labels"); pe.add_argument("--cluster"); pe.add_argument("--out")
+    pe.set_defaults(func=_cmd_evaluate)
+
+    pv = sub.add_parser("env", help="per-method environment recipes")
+    ev = pv.add_subparsers(dest="env_cmd", required=True)
+    ev.add_parser("status").set_defaults(func=_cmd_env)
+    ev.add_parser("groups").set_defaults(func=_cmd_env)
+    er = ev.add_parser("recipe"); er.add_argument("method"); er.add_argument("--name"); er.set_defaults(func=_cmd_env)
+    ey = ev.add_parser("yml"); ey.add_argument("method"); ey.add_argument("--name"); ey.add_argument("--out"); ey.set_defaults(func=_cmd_env)
+    ec = ev.add_parser("create"); ec.add_argument("method"); ec.add_argument("--name"); ec.add_argument("--run", action="store_true"); ec.set_defaults(func=_cmd_env)
+    ep = ev.add_parser("plan"); ep.add_argument("--category"); ep.add_argument("--methods", help="comma-separated method names; only their envs"); ep.set_defaults(func=_cmd_env)
+    eg = ev.add_parser("create-group"); eg.add_argument("group"); eg.add_argument("--run", action="store_true"); eg.set_defaults(func=_cmd_env)
+    edoc = ev.add_parser("doctor", help="preflight: which envs are present / need building"); edoc.add_argument("--category"); edoc.add_argument("--methods", help="comma-separated method names; only their envs"); edoc.set_defaults(func=_cmd_env)
+    ei = ev.add_parser("install", help="build every needed env from its lockfile"); ei.add_argument("--category"); ei.add_argument("--methods", help="comma-separated method names; only their envs"); ei.add_argument("--packed", action="store_true", help="use prebuilt archives when published; fall back to the lockfile build"); ei.add_argument("--run", action="store_true"); ei.set_defaults(func=_cmd_env)
+    ef = ev.add_parser("freeze", help="capture an env (or --all) to a committed lockfile"); ef.add_argument("env", nargs="?"); ef.add_argument("--all", action="store_true"); ef.add_argument("--category"); ef.set_defaults(func=_cmd_env)
+    return p
+
+
+def main(argv=None) -> int:
+    args = build_parser().parse_args(argv)
+    try:
+        return args.func(args)
+    except KeyboardInterrupt:
+        print("interrupted", file=sys.stderr)
+        return 130
+    except Exception as e:  # noqa: BLE001 - the CLI's single error boundary
+        if os.environ.get("MULTIBENCH_DEBUG"):
+            raise
+        msg = e.args[0] if isinstance(e, KeyError) and e.args else e
+        print(f"error: {msg}", file=sys.stderr)
+        print("(set MULTIBENCH_DEBUG=1 for the full traceback)", file=sys.stderr)
+        return 1
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())