nucleobench 0.1.0__tar.gz

nucleobench-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 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.

nucleobench-0.1.0/PKG-INFO

@@ -0,0 +1,82 @@
+Metadata-Version: 2.4
+Name: nucleobench
+Version: 0.1.0
+Summary: Optimizers from the nucleobench package.
+Author-email: Move37 Labs <joel.shor@move37labs.com>
+License: MIT
+Project-URL: Homepage, https://github.com/move37-labs/nucleobench
+Project-URL: Bug Tracker, https://github.com/move37-labs/nucleobench/issues
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy
+Requires-Dist: scipy
+Requires-Dist: pandas
+Requires-Dist: h5py
+Requires-Dist: pyarrow
+Requires-Dist: torch
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: coverage; extra == "dev"
+Requires-Dist: pyclean; extra == "dev"
+Dynamic: license-file
+
+# NucleoBench
+
+This is the initial repo for an upcoming paper, `NucleoBench: A Large-Scale Benchmark of Neural Nucleic Acid Design Algorithms`.
+
+This repo is covered by the MIT license.
+
+This repo is intended to be used in a few ways:
+
+1. Reproducing the results from our paper.
+1. Running the NucleoBench sequence designers on custom problems.
+1. Using our new designer, AdaBeam, on a custom problem.
+
+To do these, you can clone this repo, use the Docker image (for the benchmark), or use the PyPi package for our designers.
+
+## Results
+
+![Summary of results.](assets/images/results_summary.png)
+
+## Installation & testing
+
+Once this repo is cloned, you can make the conda/mamba/micromamba environment with:
+
+```bash
+conda env create -f environment.yml
+conda activate nucleobench
+```
+
+To test that you've install NucleoBench, run all the unittests:
+
+```bash
+pytest nucleobench/
+```
+
+You can also run the integration tests, which require an internet connection:
+
+```bash
+pytest docker_entrypoint_test.py
+```
+
+Nucleic acid design benchmark.
+
+## Running NucleoBench
+
+See the folder `recipes` for examples of how to run the designer locally.
+
+## Building a Docker image
+
+To help deploy NucleoBench to the cloud, we've created a docker container. To build it yourself, see the top of `Dockerfile` for instructions. One way of creating a docker file is:
+
+```bash
+docker build -t nucleobench -f Dockerfile .
+```

nucleobench-0.1.0/README.md

@@ -0,0 +1,52 @@
+# NucleoBench
+
+This is the initial repo for an upcoming paper, `NucleoBench: A Large-Scale Benchmark of Neural Nucleic Acid Design Algorithms`.
+
+This repo is covered by the MIT license.
+
+This repo is intended to be used in a few ways:
+
+1. Reproducing the results from our paper.
+1. Running the NucleoBench sequence designers on custom problems.
+1. Using our new designer, AdaBeam, on a custom problem.
+
+To do these, you can clone this repo, use the Docker image (for the benchmark), or use the PyPi package for our designers.
+
+## Results
+
+![Summary of results.](assets/images/results_summary.png)
+
+## Installation & testing
+
+Once this repo is cloned, you can make the conda/mamba/micromamba environment with:
+
+```bash
+conda env create -f environment.yml
+conda activate nucleobench
+```
+
+To test that you've install NucleoBench, run all the unittests:
+
+```bash
+pytest nucleobench/
+```
+
+You can also run the integration tests, which require an internet connection:
+
+```bash
+pytest docker_entrypoint_test.py
+```
+
+Nucleic acid design benchmark.
+
+## Running NucleoBench
+
+See the folder `recipes` for examples of how to run the designer locally.
+
+## Building a Docker image
+
+To help deploy NucleoBench to the cloud, we've created a docker container. To build it yourself, see the top of `Dockerfile` for instructions. One way of creating a docker file is:
+
+```bash
+docker build -t nucleobench -f Dockerfile .
+```

nucleobench-0.1.0/nucleobench/common/argparse_lib.py

@@ -0,0 +1,64 @@
+"""Utilities for parsing arguments."""
+
+from typing import Iterable
+
+import argparse
+import dataclasses
+
+@dataclasses.dataclass
+class ParsedArgs:
+    main_args: argparse.Namespace
+    model_init_args: argparse.Namespace
+    opt_init_args: argparse.Namespace
+    opt_run_args: argparse.Namespace
+    
+    
+def parse_long_start_sequence(known_args: argparse.Namespace) -> argparse.Namespace:
+    """Parse a long start sequence from a file."""
+    assert known_args.seed_sequence.startswith('local://')
+    local_fileloc = known_args.seed_sequence[len('local://'):]
+    with open(local_fileloc, 'r') as f:
+        known_args.seed_sequence = f.read()
+    return known_args
+
+
+def possibly_parse_positions_to_mutate(known_args: argparse.Namespace) -> argparse.Namespace:
+    """Possibly parse `positions_to_mutate` from a file, or leave it untouched, depending on the value."""
+    if isinstance(known_args.positions_to_mutate, str) and known_args.positions_to_mutate.startswith('local://'):
+        local_fileloc = known_args.positions_to_mutate[len('local://'):]
+        with open(local_fileloc, 'r') as f:
+            loc_str = f.read()
+        known_args.positions_to_mutate = [int(x) for x in loc_str.split('\n')]
+    elif known_args.positions_to_mutate is None or known_args.positions_to_mutate == '' or known_args.positions_to_mutate == []:
+        known_args.positions_to_mutate = None
+    else:
+        assert isinstance(known_args.positions_to_mutate, list), (type(known_args.positions_to_mutate), known_args.positions_to_mutate)
+        known_args.positions_to_mutate = [int(x) for x in known_args.positions_to_mutate.split(',')]
+    return known_args
+
+
+def handle_leftover_args(known_args: argparse.Namespace, leftover_args: Iterable):
+    """Handle leftover arguments, either by failing or by ignoring them."""
+    if known_args.ignore_empty_cmd_args:
+        # Check that every "value" is either `None` or `empty`. If so, allow it to continue.
+        for i in leftover_args:
+            if i.startswith('--'):
+                if '=' in i:
+                    arg_val = i.split('=')[1]
+                    if arg_val not in [None, '']:
+                        raise ValueError(f'Unused arg, not empty: {leftover_args}')
+                continue
+            else: 
+                if i not in [None, '']:
+                    raise ValueError(f'Unused arg, not empty: {leftover_args}')
+    else:
+        raise ValueError(f'Unused args: {leftover_args}')
+    
+
+def str_to_bool(s):
+    if s.lower() in ('yes', 'true', 't', '1'):
+        return True
+    elif s.lower() in ('no', 'false', 'f', '0'):
+        return False
+    else:
+        raise argparse.ArgumentTypeError('Boolean value expected.')

nucleobench-0.1.0/nucleobench/common/attribution_lib_torch.py

@@ -0,0 +1,199 @@
+"""Library for smoothgrad and genome-specific attribution methods.
+
+Ref:
+1. [Correcting gradient-based interpretations of deep neural networks for genomics](https://genomebiology.biomedcentral.com/articles/10.1186/s13059-023-02956-3)
+2. [SmoothGrad: removing noise by adding noise](https://arxiv.org/abs/1706.03825)
+3. [Quick and effective approximation of in silico saturation mutagenesis experiments with first-order taylor expansion](https://pubmed.ncbi.nlm.nih.gov/39286491/)
+
+TODO(joelshor): Consider using public version of attribution tools, such as:
+- PyTorch Smoothgrad: https://github.com/pkmr06/pytorch-smoothgrad
+- PyTorch GradCam and others: https://github.com/jacobgil/pytorch-grad-cam?tab=readme-ov-file
+- PyTorch Smoothgrad and others: https://tf-explain.readthedocs.io/en/latest/
+
+To test locally:
+```zsh
+python -m nucleobench.common.attribution_lib
+```
+"""
+
+import gc
+import numpy as np
+import torch
+from typing import Callable, Optional
+
+
+TISMOutputType = list[dict[str, float]]
+SmoothgradVocabType = list[dict[str, torch.Tensor]]
+TISMLocationsType = list[int]
+
+
+def noise_inputs(
+    input_tensor: torch.Tensor,
+    noise_stdev: float,
+    times: int,
+    ) -> torch.Tensor:
+    """Generates noisy inputs.
+
+    NOTE: For simplicity, for now, we work with SINGLE TENSORS. Assume no batch dimension.
+
+    Args:
+        input_tensor: Input tensor. Doesn't have to be genomic. Should NOT be batched.
+        noise_stdev: Noise to add.
+        times: Number of times to add noise.
+    """
+    if noise_stdev < 0:
+        raise ValueError(f'Requires non-negative noise stdev: {noise_stdev}')
+    x = input_tensor  # Syntactic sugar.
+    
+    # Stack N versions of the input, to add uncorrelated noise to.
+    with torch.no_grad():
+        x = x.unsqueeze(0)
+        x = x.repeat([times] + [1] * (x.ndim-1))
+
+        # Add noise for hte smoothgrad algorithm.
+        if noise_stdev > 0:
+            noise_to_add = torch.normal(mean=torch.zeros(x.shape), std=noise_stdev)
+            x += noise_to_add
+    return x
+
+def noisy_grads_torch(
+        input_tensor: torch.Tensor, 
+        model: Callable[[torch.Tensor], torch.Tensor], 
+        noise_stdev: float, 
+        times: int,
+        idxs: Optional[TISMLocationsType] = None,
+        ) -> torch.Tensor:
+    """Generates noisy gradients from a function.
+
+    NOTE: For simplicity, for now, we work with SINGLE TENSORS. Assume no batch dimension.
+
+    This replicates the input `times` times, and runs it through the network all at once.
+
+    TODO(joelshor): Add batching, for the situation where `times` is larger than the possible batch size
+        of a single inference with a network.
+    TODO(joelshor): Add ability to efficiently compute multiple inputs at once.
+
+    Args:
+        input_tensor (torch.Tensor): Input tensor. Doesn't have to be genomic. Should NOT be batched.
+        model: PyTorch model to use. The model must return a scalar per batch element.
+        noise_stdev: Noise to add.
+        times: Number of times to add noise.
+        idx: If present, only backprop through this location.
+    """
+    x = noise_inputs(
+        input_tensor=input_tensor,
+        noise_stdev=noise_stdev,
+        times=times)
+
+    # Run inference to get grads.
+    if idxs is None:
+        x_grad = x
+        x_grad.requires_grad = True
+    else:
+        x, x_grad = apply_gradient_mask(x, idxs)
+        
+    y = model(x)
+    y_sum = y.sum()
+    y_sum.backward(retain_graph=False)
+    noisy_grads = x_grad.grad.numpy()
+
+    gc.collect()
+    torch.cuda.empty_cache()
+
+    assert noisy_grads.shape == x_grad.shape
+    return noisy_grads
+
+
+def smoothgrad_torch(
+    input_tensor: torch.Tensor, 
+    model: Callable[[torch.Tensor], torch.Tensor], 
+    noise_stdev: float, 
+    times: int,
+    idxs: Optional[TISMLocationsType] = None,
+    ) -> torch.Tensor:
+    """Custom implementation of SmoothGrad.
+    https://arxiv.org/pdf/1706.03825
+
+    NOTE: For simplicity, for now, we work with SINGLE TENSORS. Assume no batch dimension.
+
+    This replicates the input `times` times, and runs it through the network all at once.
+
+    TODO(joelshor): Add batching, for the situation where `times` is larger than the possible batch size
+        of a single inference with a network.
+    TODO(joelshor): Add ability to efficiently compute multiple inputs at once.
+
+    Args:
+        input_tensor (torch.Tensor): Input tensor. Doesn't have to be genomic. Should NOT be batched.
+        model: PyTorch model to use. The model must return a scalar per batch element.
+        noise_stdev: Noise to add.
+        times: Number of times to add noise.
+        idx: If present, only backprop to this location.
+        
+    Returns:
+        Per-nucleotide smoothgrad.
+    """
+    noisy_grads = noisy_grads_torch(
+        input_tensor=input_tensor,
+        model=model,
+        noise_stdev=noise_stdev,
+        times=times,
+        idxs=idxs,
+    )
+
+    return np.mean(noisy_grads, axis=0)
+
+
+# TODO(joelshor): Add `attribution_lib.py` test, taken from `malinois/model_def_test.py`.
+def smoothgrad_tensor_to_dict(smooth_grad: torch.Tensor, vocab: list[str]) -> SmoothgradVocabType:
+    """Map the smoothgrad indices to the vocab."""
+    assert smooth_grad.ndim == 2
+    assert list(smooth_grad.shape)[0] == len(vocab)
+    def _to_dict(x: torch.Tensor) -> dict[str, torch.Tensor]:
+        return {vocab[i]: x[i] for i in range(len(vocab))}
+    return [_to_dict(x) for x in smooth_grad.T]
+
+
+def smoothgrad_to_tism(sg: SmoothgradVocabType, base_seq: str) -> TISMOutputType:
+    """Returns result according to Taylor in-silico mutagenesis.
+    
+    Paper: https://www.cell.com/iscience/fulltext/S2589-0042(24)02032-7"""
+    assert len(sg) == len(base_seq)
+
+    tism = []
+    for base_nt, sg_dict in zip(base_seq, sg):
+        cur_tism = {}
+        for nt, sg in sg_dict.items():
+            if nt == base_nt: continue
+            cur_tism[nt] = float(sg - sg_dict[base_nt])
+        tism.append(cur_tism)
+
+    return tism
+
+
+def apply_gradient_mask(x: torch.Tensor, idxs: TISMLocationsType) -> tuple[torch.Tensor, torch.Tensor]:
+    """Applies a gradient mask to the input tensor.
+    
+    NOTE: Do NOT just multiply by 0. This will run out of memory in large models.
+
+    Returns:
+        Tuple of (x, masked_x), where masked_x is the input tensor with the gradient mask applied.
+    """
+    assert min(idxs) >= 0
+    assert max(idxs) < x.shape[2]
+    assert x.ndim == 3, x.shape
+    
+    no_gradient = x.clone().detach()
+    no_gradient.requires_grad = False
+    
+    x_grad = x[:, :, idxs].clone().detach()
+    x_grad.requires_grad = True
+    x_grad_i = {idx: i for i, idx in enumerate(idxs)}
+    
+    # Instead of using `torch.where`, we use this method to make our gradient tensor
+    # as small as possible, to preserve memory.
+    tensor_slices = [x_grad[:, :, x_grad_i[i]:x_grad_i[i]+1] if i in idxs
+                     else no_gradient[:, :, i:i+1]
+                     for i in range(no_gradient.shape[2])]
+    x = torch.concat(tensor_slices, dim=2)
+    
+    return x, x_grad