478 plex init not working in pip package #479

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged

thetechnocrat-dev merged 6 commits into main from 478-plex-init-not-working-in-pip-package

Jul 13, 2023

cmd/vectorize.go

-Original file line number
+Diff line change
@@ Expand Up @@
     						}
     					}
+    					cidPath := fmt.Sprintf("%s/%s", fileOutput.IPFS, fileOutput.FilePath)
     					ov.FilePaths = append(ov.FilePaths, absoluteFilePath)
     					ov.CIDs = append(ov.CIDs, fileOutput.IPFS)
+    					ov.CidPaths = append(ov.CidPaths, cidPath)
     					outputMap[key] = ov
     				}
     			}
@@ Expand Down @@

internal/ipwl/io.go

-Original file line number
+Diff line change
@@ Expand Up / @@ -157,4 +157,5 @@ func PrintIOGraphStatus(ioList []IO) { @@
     type OutputValues struct {
     	FilePaths []string `json:"filePaths"`
     	CIDs      []string `json:"cids"`
+    	CidPaths  []string `json:"cidPaths"`
     }

python/README.md

-Original file line number
+Diff line change
@@ Expand Up / @@ -23,3 +23,10 @@ export PLAT_NAME=win_amd64 @@
     python setup.py bdist_wheel --plat-name win_amd64
     twine upload dist/*
     ```
+    # Dev Setups
+    ```
+    cd python/dev
+    pip install -e ../
+    python example.py
+    ```

python/dev/example.py

-Original file line number
+Diff line change
@@ -0,0 +1,30 @@
+    import os
+    from plex import CoreTools, ScatteringMethod, plex_init, plex_run, plex_vectorize, plex_mint
+    plex_dir = os.path.dirname(os.path.dirname(os.getcwd()))
+    plex_path = os.path.join(plex_dir, "plex")
+    jobs_dir = os.path.join(plex_dir, "jobs")
+    test_data_dir = os.path.join(plex_dir, "testdata")
+    print(f"Using plex_path, {plex_path}, if this looks incorrect then make sure you are running from the python/dev directory")
+    small_molecules = [f"{test_data_dir}/binding/abl/ZINC000003986735.sdf", f"{test_data_dir}/binding/abl/ZINC000019632618.sdf"]
+    proteins = [f"{test_data_dir}/binding/abl/7n9g.pdb"]
+    initial_io_cid = plex_init(
+        CoreTools.EQUIBIND.value,
+        ScatteringMethod.CROSS_PRODUCT.value,
+        plex_path=plex_path,
+        small_molecule=small_molecules,
+        protein=proteins)
+    completed_io_cid, io_file_path = plex_run(initial_io_cid, output_dir=jobs_dir, plex_path=plex_path)
+    vectors = plex_vectorize(io_file_path, CoreTools.EQUIBIND.value, plex_path=plex_path)
+    print(vectors)
+    print(vectors['best_docked_small_molecule']['filePaths'])
+    print(vectors['best_docked_small_molecule']['cids'])
+    plex_mint(completed_io_cid, plex_path=plex_path)

python/dev/rfdiffusion.py

-Original file line number
+Diff line change
@@ -0,0 +1,125 @@
+    import os
+    import json
+    import requests
+    from tempfile import TemporaryDirectory, NamedTemporaryFile
+    from plex import CoreTools, ScatteringMethod, plex_init, plex_run, plex_vectorize, plex_mint, plex_upload
+    plex_dir = os.path.dirname(os.path.dirname(os.getcwd()))
+    plex_path = os.path.join(plex_dir, "plex")
+    jobs_dir = os.path.join(plex_dir, "jobs")
+    def move_from_file_url_to_ipfs(url: str) -> str:
+        # Create a temporary directory using the `with` statement, so it's automatically cleaned up when we're done
+        with TemporaryDirectory() as tmp_dir:
+            # Get the file name from the url
+            file_name = url.split('/')[-1]
+            # Create the path to save the file
+            save_path = os.path.join(tmp_dir, file_name)
+            # Send a HTTP request to the url
+            response = requests.get(url, stream=True)
+            if response.status_code == 200:
+                # If the request is successful, open the file in write mode and download the file
+                with open(save_path, 'wb') as f:
+                    for chunk in response.iter_content(chunk_size=1024):
+                        if chunk:
+                            f.write(chunk)
+                print(f"File downloaded successfully and saved as {save_path}")
+                cid = plex_upload(save_path, plex_path=plex_path)
+            else:
+                print(f"Failed to download file. HTTP Response Code: {response.status_code}")
+                cid = ""
+            return cid, file_name
+    url = 'https://raw.githubusercontent.com/labdao/plex/447-add-rfdiffusion-to-plex/tools/rfdiffusion/6vja_stripped.pdb'
+    protein_6vja_cid, protein_6vja_file_name = move_from_file_url_to_ipfs(url)
+    hotspot_rfdiffusion_tool = {
+        "class": "CommandLineTool",
+        "name": "rfdiffusion_hotspot",
+        "description": "design protein binders; generally useful for conditional generation of protein backbones",
+        "baseCommand": ["/bin/bash", "-c"],
+        "arguments": [
+            "source activate SE3nv && python3 /app/scripts/run_inference.py 'contigmap.contigs=[$(inputs.motif.default) $(inputs.binder_length_min.default)-$(inputs.binder_length_max.default)]' $(inputs.hotspot.default) inference.input_pdb=$(inputs.protein.filepath) inference.output_prefix=/outputs/$(inputs.protein.basename)_backbone inference.num_designs=$(inputs.number_of_designs.default) denoiser.noise_scale_ca=0 denoiser.noise_scale_frame=0;"
+          ],
+        "dockerPull": "public.ecr.aws/p7l9w5o7/rfdiffusion:latest@sha256:0a6ff53004958ee5e770b0b25cd7f270eaf9fc285f6e91f17ad4024d2cc4ea91",
+        "gpuBool": True,
+        "networkBool": False,
+        "inputs": {
+          "protein": {
+            "type": "File",
+            "item": "",
+            "glob": ["*.pdb"]
+          },
+          "motif": {
+            "type": "string",
+            "item": "",
+            "default": "D46-200/0"
+          },
+          "hotspot": {
+            "type": "string",
+            "item": "",
+            "default": "'ppi.hotspot_res=[D170, D171, D172, D173, D174, D76, D161]'"
+          },
+          "binder_length_min": {
+            "type": "int",
+            "item": "",
+            "default": "50"
+          },
+          "binder_length_max": {
+            "type": "int",
+            "item": "",
+            "default": "100"
+          },
+          "number_of_designs": {
+            "type": "int",
+            "item": "",
+            "default": "10"
+          }
+        },
+        "outputs": {
+          "designed_backbones": {
+            "type": "Array",
+            "item": "File",
+            "glob": ["*_backbone_*.pdb"]
+          },
+          "first_designed_backbone": {
+            "type": "File",
+            "item": "",
+            "glob": ["*_backbone_0.pdb"]
+          }
+        }
+    }
+    with NamedTemporaryFile(suffix=".json", delete=False, mode='w') as temp_file:
+        # Use json.dump to write the dictionary to the file
+        json.dump(hotspot_rfdiffusion_tool, temp_file)
+        print(f"Temporary file saved as {temp_file.name}")
+        # flush data to disk
+        temp_file.flush()
+        hotspot_rfdiffusion_tool_cid = plex_upload(temp_file.name, wrap_file=False, plex_path=plex_path)
+        print(f"Tool saved to IPFS as {hotspot_rfdiffusion_tool_cid}")
+    initial_io_cid = plex_init(
+        hotspot_rfdiffusion_tool_cid,
+        plex_path=plex_path,
+        protein=[f'{protein_6vja_cid}/{protein_6vja_file_name}'],
+        )
+    CACHE = True
+    if CACHE:
+        completed_io_cid = 'QmbHneUZZpNh24is1uTnCpMYKRWStgEqAaYL4aDjN6tNzQ'
+        io_cid_file_path = ''
+    else:
+        completed_io_cid, io_cid_file_path = plex_run(initial_io_cid, plex_path=plex_path)
+    vectors = plex_vectorize(completed_io_cid, hotspot_rfdiffusion_tool_cid, plex_path=plex_path)
+    print(vectors['first_designed_backbone']['cidPaths'])

python/setup.py

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -16,7 +16,7 @@ def run(self):
  
            # Retrieve platform from environment variable

            plat_name = os.environ['PLAT_NAME']

            current_binary_version = "0.8.2"

            current_binary_version = "0.8.3"

            # map plat_name to go_bin_url

            urls = {

                "darwin_x86_64": f"https://github.com/labdao/plex/releases/download/v{current_binary_version}/plex_{current_binary_version}_darwin_amd64.tar.gz",

    @@ -54,7 +54,7 @@ def download_and_extract(self, go_bin_url, temp_dir):
  
    setup(

        name="PlexLabExchange",

        version="0.8.16",

        version="0.8.17",

        packages=find_packages(where='src'),  # tell setuptools to look in the 'src' directory for packages

        package_dir={'': 'src'},  # tell setuptools that the packages are under the 'src' directory

        cmdclass={

python/src/plex/__init__.py

            
                      Original file line number
                      Diff line number
                      Diff line change
                  
    @@ -7,8 +7,9 @@
  
    class ScatteringMethod(Enum):

        DOT_PRODUCT = 'dotproduct'

        CROSS_PRODUCT = 'cross_product'

        # values are camel case to match GoLang CLI

        DOT_PRODUCT = 'dotProduct'

        CROSS_PRODUCT = 'crossProduct'

    class CoreTools(Enum):

    @@ -35,35 +36,39 @@ def __init__(self, message):
  
            super().__init__(f"{self.message}\n{self.github_issue_message}")

    def plex_init(toolpath: str, scatteringMethod="dotProduct", plex_path="plex", **kwargs):

    def plex_init(tool_path: str, scattering_method=ScatteringMethod.DOT_PRODUCT.value, plex_path="plex", **kwargs):

        cwd = os.getcwd()

        plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(os.path.dirname(cwd)))

        # Convert kwargs dictionary to a JSON string

        inputs = json.dumps(kwargs)

        cmd = [plex_path, "init", "-t", toolpath, "-i", inputs, f"--scatteringMethod={scatteringMethod}"]

        cmd = [plex_path, "init", "-t", tool_path, "-i", inputs, f"--scatteringMethod={scattering_method}"]

        print(' '.join(cmd))

        io_json_cid = ""

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

            for line in p.stdout:

                if "Pinned IO JSON CID:" in line:

                    parts = line.split()

                    io_json_cid = parts[-1]

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

        if io_json_cid == "":

            raise PlexError("Failed to initialize IO JSON CID")

        return io_json_cid

    def plex_vectorize(io_path: str, tool_cid: str, outputDir="", plex_path="plex"):

    def plex_vectorize(io_path: str, tool_cid: str, output_dir="", plex_path="plex"):

        cwd = os.getcwd()

        plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(os.path.dirname(cwd)))

        cmd = [plex_path, "vectorize", "-i", io_path, "-t", tool_cid, "-o", outputDir]

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

        cmd = [plex_path, "vectorize", "-i", io_path, "-t", tool_cid, "-o", output_dir]

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

            outvects = ""

            for line in p.stdout:

                if "Output Vectors were saved at:" in line:

    @@ -73,40 +78,44 @@ def plex_vectorize(io_path: str, tool_cid: str, outputDir="", plex_path="plex"):
  
                        outvects = json.load(f)

                    os.remove(io_vector_outpath)

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

        if outvects == "":

            raise PlexError("Failed to vectorize IO JSON CID")

        return outvects

    def plex_upload(filePath: str, wrapFile=True, plex_path="plex"):

        cmd = [plex_path, "upload", "-p", filePath]

    def plex_upload(file_path: str, wrap_file=True, plex_path="plex"):

        cmd = [plex_path, "upload", "-p", file_path]

        if not wrapFile:

        if not wrap_file:

            cmd.append("-w=false")

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True) as p:

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True) as p:

            file_cid = ""

            for line in p.stdout:

                if "Uploaded CID:" in line:

                    parts = line.split()

                    file_cid = parts[-1]

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

        if file_cid == "":

            raise PlexError("Failed to upload file to IPFS")

        return file_cid

    def plex_create(toolpath: str, inputDir: str, layers=2, outputDir="", verbose=False, showAnimation=False, concurrency="1", annotations=[], plex_path="plex"):

    def plex_create(tool_path: str, input_dir: str, layers=2, output_dir="", verbose=False, show_animation=False, concurrency="1", annotations=[], plex_path="plex"):

        cwd = os.getcwd()

        plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(os.path.dirname(cwd)))

        cmd = [plex_path, "create", "-t", toolpath, "-i", inputDir, f"--layers={layers}"]

        cmd = [plex_path, "create", "-t", tool_path, "-i", input_dir, f"--layers={layers}"]

        if outputDir:

            cmd.append(f"-o={outputDir}")

        if output_dir:

            cmd.append(f"-o={output_dir}")

        if verbose:

            cmd.append("-v=true")

    @@ -117,31 +126,33 @@ def plex_create(toolpath: str, inputDir: str, layers=2, outputDir="", verbose=Fa
  
        if annotations:

            cmd.append(f"--annotations={annotations.join(',')}")

        if not showAnimation: # default is true in the CLI

        if not show_animation: # default is true in the CLI

            cmd.append("--showAnimation=false")

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

            io_json_cid = ""

            for line in p.stdout:

                if "Initial IO JSON file CID:" in line:

                    parts = line.split()

                    io_json_cid = parts[-1]

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

        if io_json_cid == "":

            raise PlexError("Failed to create IO JSON CID")

        return io_json_cid

    def plex_run(io_json_cid: str, outputDir="", verbose=False, showAnimation=False, concurrency="1", annotations=[], plex_path="plex"):

    def plex_run(io_json_cid: str, output_dir="", verbose=False, show_animation=False, concurrency="1", annotations=[], plex_path="plex"):

        cwd = os.getcwd()

        # plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(os.path.dirname(cwd)))

        plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(cwd))

        cmd = [plex_path, "run", "-i", io_json_cid]

        if outputDir:

            cmd.append(f"-o={outputDir}")

        if output_dir:

            cmd.append(f"-o={output_dir}")

        if verbose:

            cmd.append("-v=true")

    @@ -152,10 +163,10 @@ def plex_run(io_json_cid: str, outputDir="", verbose=False, showAnimation=False,
  
        if annotations:

            cmd.append(f"--annotations={annotations.join(',')}")

        if not showAnimation: # default is true in the CLI

        if not show_animation: # default is true in the CLI

            cmd.append("--showAnimation=false")

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

            io_json_cid = ""

            io_json_local_filepath = ""

            for line in p.stdout:

    @@ -166,24 +177,28 @@ def plex_run(io_json_cid: str, outputDir="", verbose=False, showAnimation=False,
  
                    parts = line.split()

                    io_json_local_filepath = parts[-1]

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

        if io_json_cid == "" or io_json_local_filepath == "":

            raise PlexError("Failed to run IO JSON CID")

        return io_json_cid, io_json_local_filepath

    def plex_mint(io_json_cid: str, imageCid="", plex_path="plex"):

    def plex_mint(io_json_cid: str, image_cid="", plex_path="plex"):

        cwd = os.getcwd()

        plex_work_dir = os.environ.get("PLEX_WORK_DIR", os.path.dirname(os.path.dirname(cwd)))

        cmd = [plex_path, "mint", "-i", io_json_cid]

        if imageCid:

            cmd.append(f"-imageCid={imageCid}")

        if image_cid:

            cmd.append(f"-imageCid={image_cid}")

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

        with subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, universal_newlines=True, cwd=plex_work_dir) as p:

            for line in p.stdout:

                print(line, end='')

            for line in p.stderr:

                print(line, end='')

    def print_io_graph_status(io_graph):

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

478 plex init not working in pip package #479

Uh oh!

Diff view

Diff view

There are no files selected for viewing