Merge pull request #100 from biggraph/darabos-generative-screening-skeleton

examples/Generative drug screening

@@ -0,0 +1,873 @@
+{
+  "edges": [
+    {
+      "id": "Import file 2 Query GenMol 1",
+      "source": "Import file 2",
+      "sourceHandle": "output",
+      "target": "Query GenMol 1",
+      "targetHandle": "bundle"
+    },
+    {
+      "id": "Import file 1 MSA-search 1",
+      "source": "Import file 1",
+      "sourceHandle": "output",
+      "target": "MSA-search 1",
+      "targetHandle": "bundle"
+    },
+    {
+      "id": "Query GenMol 1 Query DiffDock 1",
+      "source": "Query GenMol 1",
+      "sourceHandle": "output",
+      "target": "Query DiffDock 1",
+      "targetHandle": "ligands"
+    },
+    {
+      "id": "Query DiffDock 1 View molecules 1",
+      "source": "Query DiffDock 1",
+      "sourceHandle": "output",
+      "target": "View molecules 1",
+      "targetHandle": "bundle"
+    },
+    {
+      "id": "MSA-search 1 Query OpenFold2 1",
+      "source": "MSA-search 1",
+      "sourceHandle": "output",
+      "target": "Query OpenFold2 1",
+      "targetHandle": "bundle"
+    },
+    {
+      "id": "Query OpenFold2 1 View molecules 3",
+      "source": "Query OpenFold2 1",
+      "sourceHandle": "output",
+      "target": "View molecules 3",
+      "targetHandle": "bundle"
+    },
+    {
+      "id": "Query OpenFold2 1 Query DiffDock 1",
+      "source": "Query OpenFold2 1",
+      "sourceHandle": "output",
+      "target": "Query DiffDock 1",
+      "targetHandle": "proteins"
+    }
+  ],
+  "env": "LynxKite Graph Analytics",
+  "nodes": [
+    {
+      "data": {
+        "__execution_delay": 0.0,
+        "collapsed": null,
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {},
+          "name": "Import file",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "file_format": {
+              "default": "csv",
+              "groups": {
+                "csv": [
+                  {
+                    "default": "<from file>",
+                    "name": "columns",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  },
+                  {
+                    "default": "<auto>",
+                    "name": "separator",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  }
+                ],
+                "excel": [
+                  {
+                    "default": "Sheet1",
+                    "name": "sheet_name",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  }
+                ],
+                "json": [],
+                "parquet": []
+              },
+              "name": "file_format",
+              "selector": {
+                "default": "csv",
+                "name": "file_format",
+                "type": {
+                  "enum": [
+                    "csv",
+                    "parquet",
+                    "json",
+                    "excel"
+                  ]
+                }
+              },
+              "type": "group"
+            },
+            "file_path": {
+              "default": null,
+              "name": "file_path",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "table_name": {
+              "default": null,
+              "name": "table_name",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            }
+          },
+          "type": "basic"
+        },
+        "params": {
+          "columns": "<from file>",
+          "file_format": "csv",
+          "file_path": "uploads/protein.csv",
+          "separator": "<auto>",
+          "table_name": ""
+        },
+        "status": "done",
+        "title": "Import file"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 487.0,
+      "id": "Import file 1",
+      "position": {
+        "x": -755.0582906538923,
+        "y": 543.770372030674
+      },
+      "type": "basic",
+      "width": 439.0
+    },
+    {
+      "data": {
+        "__execution_delay": 0.0,
+        "collapsed": null,
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {},
+          "name": "Import file",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "file_format": {
+              "default": "csv",
+              "groups": {
+                "csv": [
+                  {
+                    "default": "<from file>",
+                    "name": "columns",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  },
+                  {
+                    "default": "<auto>",
+                    "name": "separator",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  }
+                ],
+                "excel": [
+                  {
+                    "default": "Sheet1",
+                    "name": "sheet_name",
+                    "type": {
+                      "type": "<class 'str'>"
+                    }
+                  }
+                ],
+                "json": [],
+                "parquet": []
+              },
+              "name": "file_format",
+              "selector": {
+                "default": "csv",
+                "name": "file_format",
+                "type": {
+                  "enum": [
+                    "csv",
+                    "parquet",
+                    "json",
+                    "excel"
+                  ]
+                }
+              },
+              "type": "group"
+            },
+            "file_path": {
+              "default": null,
+              "name": "file_path",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "table_name": {
+              "default": null,
+              "name": "table_name",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            }
+          },
+          "type": "basic"
+        },
+        "params": {
+          "columns": "<from file>",
+          "file_format": "csv",
+          "file_path": "uploads/molecules.csv",
+          "separator": "<auto>",
+          "table_name": null
+        },
+        "status": "done",
+        "title": "Import file"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 436.0,
+      "id": "Import file 2",
+      "position": {
+        "x": 62.887657256500006,
+        "y": 1380.6697994924546
+      },
+      "type": "basic",
+      "width": 311.0
+    },
+    {
+      "data": {
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {
+            "bundle": {
+              "name": "bundle",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "Query GenMol",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "molecule_column": {
+              "default": null,
+              "name": "molecule_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "molecule_table": {
+              "default": null,
+              "name": "molecule_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "noise": {
+              "default": 0.2,
+              "name": "noise",
+              "type": {
+                "type": "<class 'float'>"
+              }
+            },
+            "num_molecules": {
+              "default": 5.0,
+              "name": "num_molecules",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            },
+            "scoring": {
+              "default": "QED",
+              "name": "scoring",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "step_size": {
+              "default": 4.0,
+              "name": "step_size",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            },
+            "temperature": {
+              "default": 1.0,
+              "name": "temperature",
+              "type": {
+                "type": "<class 'float'>"
+              }
+            }
+          },
+          "position": {
+            "x": 594.0,
+            "y": 633.0
+          },
+          "type": "basic"
+        },
+        "params": {
+          "molecule_column": null,
+          "molecule_table": null,
+          "noise": 0.2,
+          "num_molecules": 5.0,
+          "scoring": "QED",
+          "step_size": 4.0,
+          "temperature": 1.0
+        },
+        "status": "done",
+        "title": "Query GenMol"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 601.0,
+      "id": "Query GenMol 1",
+      "position": {
+        "x": 663.3333333333335,
+        "y": 1283.3333333333335
+      },
+      "type": "basic",
+      "width": 358.0
+    },
+    {
+      "data": {
+        "collapsed": null,
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {
+            "bundle": {
+              "name": "bundle",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "MSA-search",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "databases": {
+              "default": "[\"Uniref30_2302\", \"colabfold_envdb_202108\", \"PDB70_220313\"]",
+              "name": "databases",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "e_value": {
+              "default": 0.0001,
+              "name": "e_value",
+              "type": {
+                "type": "<class 'float'>"
+              }
+            },
+            "iterations": {
+              "default": 1.0,
+              "name": "iterations",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            },
+            "output_alignment_formats": {
+              "default": [
+                "fasta",
+                "a3m"
+              ],
+              "name": "output_alignment_formats",
+              "type": {
+                "type": "list[lynxkite_bio.nims.AlignmentFormats]"
+              }
+            },
+            "protein_column": {
+              "default": null,
+              "name": "protein_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "protein_table": {
+              "default": null,
+              "name": "protein_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "search_type": {
+              "default": "ALPHAFOLD2",
+              "name": "search_type",
+              "type": {
+                "enum": [
+                  "ALPHAFOLD2",
+                  "ESM2"
+                ]
+              }
+            }
+          },
+          "position": {
+            "x": 576.0,
+            "y": 228.0
+          },
+          "type": "basic"
+        },
+        "params": {
+          "databases": "[\"Uniref30_2302\", \"colabfold_envdb_202108\", \"PDB70_220313\"]",
+          "e_value": 0.0001,
+          "iterations": 1.0,
+          "output_alignment_formats": [
+            "fasta",
+            "a3m"
+          ],
+          "protein_column": null,
+          "protein_table": null,
+          "search_type": "ALPHAFOLD2"
+        },
+        "status": "done",
+        "title": "MSA-search"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 550.0,
+      "id": "MSA-search 1",
+      "position": {
+        "x": -45.0,
+        "y": 570.0
+      },
+      "type": "basic",
+      "width": 531.0
+    },
+    {
+      "data": {
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {
+            "ligands": {
+              "name": "ligands",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            },
+            "proteins": {
+              "name": "proteins",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "Query DiffDock",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "ligand_column": {
+              "default": null,
+              "name": "ligand_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "ligand_table": {
+              "default": null,
+              "name": "ligand_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "num_poses": {
+              "default": 10.0,
+              "name": "num_poses",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            },
+            "num_steps": {
+              "default": 18.0,
+              "name": "num_steps",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            },
+            "protein_column": {
+              "default": null,
+              "name": "protein_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "protein_table": {
+              "default": null,
+              "name": "protein_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "time_divisions": {
+              "default": 20.0,
+              "name": "time_divisions",
+              "type": {
+                "type": "<class 'int'>"
+              }
+            }
+          },
+          "position": {
+            "x": 852.0,
+            "y": 432.0
+          },
+          "type": "basic"
+        },
+        "params": {
+          "ligand_column": null,
+          "ligand_table": null,
+          "num_poses": 10.0,
+          "num_steps": 18.0,
+          "protein_column": null,
+          "protein_table": null,
+          "time_divisions": 20.0
+        },
+        "status": "done",
+        "title": "Query DiffDock"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 635.0,
+      "id": "Query DiffDock 1",
+      "position": {
+        "x": 1543.010053920781,
+        "y": 1167.386382170133
+      },
+      "type": "basic",
+      "width": 408.0
+    },
+    {
+      "data": {
+        "display": {
+          "series": [
+            {
+              "data": [
+                {
+                  "name": "Hydrogen",
+                  "value": 2
+                },
+                {
+                  "name": "Sulfur",
+                  "value": 1
+                },
+                {
+                  "name": "Oxygen",
+                  "value": 4
+                }
+              ],
+              "itemStyle": {
+                "borderColor": "#fff",
+                "borderRadius": 10,
+                "borderWidth": 2
+              },
+              "radius": [
+                "40%",
+                "70%"
+              ],
+              "type": "pie"
+            }
+          ]
+        },
+        "error": null,
+        "meta": {
+          "inputs": {
+            "bundle": {
+              "name": "bundle",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "View molecules",
+          "outputs": {},
+          "params": {
+            "color": {
+              "default": "spectrum",
+              "name": "color",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "molecule_column": {
+              "default": null,
+              "name": "molecule_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "molecule_table": {
+              "default": null,
+              "name": "molecule_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            }
+          },
+          "position": {
+            "x": 1009.0,
+            "y": 124.0
+          },
+          "type": "visualization"
+        },
+        "params": {
+          "color": "spectrum",
+          "molecule_column": null,
+          "molecule_table": null
+        },
+        "status": "done",
+        "title": "View molecules"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 200.0,
+      "id": "View molecules 3",
+      "position": {
+        "x": 1545.0,
+        "y": 585.0
+      },
+      "type": "visualization",
+      "width": 200.0
+    },
+    {
+      "data": {
+        "display": {
+          "series": [
+            {
+              "data": [
+                {
+                  "name": "Hydrogen",
+                  "value": 2
+                },
+                {
+                  "name": "Sulfur",
+                  "value": 1
+                },
+                {
+                  "name": "Oxygen",
+                  "value": 4
+                }
+              ],
+              "itemStyle": {
+                "borderColor": "#fff",
+                "borderRadius": 10,
+                "borderWidth": 2
+              },
+              "radius": [
+                "40%",
+                "70%"
+              ],
+              "type": "pie"
+            }
+          ]
+        },
+        "error": null,
+        "meta": {
+          "inputs": {
+            "bundle": {
+              "name": "bundle",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "View molecules",
+          "outputs": {},
+          "params": {
+            "color": {
+              "default": "spectrum",
+              "name": "color",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "molecule_column": {
+              "default": null,
+              "name": "molecule_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "molecule_table": {
+              "default": null,
+              "name": "molecule_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            }
+          },
+          "position": {
+            "x": 859.0,
+            "y": 225.0
+          },
+          "type": "visualization"
+        },
+        "params": {
+          "color": "spectrum",
+          "molecule_column": null,
+          "molecule_table": null
+        },
+        "status": "done",
+        "title": "View molecules"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 200.0,
+      "id": "View molecules 1",
+      "position": {
+        "x": 2230.0,
+        "y": 1598.3333333333333
+      },
+      "type": "visualization",
+      "width": 200.0
+    },
+    {
+      "data": {
+        "display": null,
+        "error": null,
+        "meta": {
+          "inputs": {
+            "bundle": {
+              "name": "bundle",
+              "position": "left",
+              "type": {
+                "type": "<class 'lynxkite_graph_analytics.core.Bundle'>"
+              }
+            }
+          },
+          "name": "Query OpenFold2",
+          "outputs": {
+            "output": {
+              "name": "output",
+              "position": "right",
+              "type": {
+                "type": "None"
+              }
+            }
+          },
+          "params": {
+            "alignment_column": {
+              "default": null,
+              "name": "alignment_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "alignment_table": {
+              "default": null,
+              "name": "alignment_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "databases": {
+              "default": "[\"Uniref30_2302\", \"colabfold_envdb_202108\", \"PDB70_220313\"]",
+              "name": "databases",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "protein_column": {
+              "default": null,
+              "name": "protein_column",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "protein_table": {
+              "default": null,
+              "name": "protein_table",
+              "type": {
+                "type": "<class 'str'>"
+              }
+            },
+            "relaxed_prediction": {
+              "default": false,
+              "name": "relaxed_prediction",
+              "type": {
+                "type": "<class 'bool'>"
+              }
+            },
+            "use_templates": {
+              "default": false,
+              "name": "use_templates",
+              "type": {
+                "type": "<class 'bool'>"
+              }
+            }
+          },
+          "position": {
+            "x": 628.0,
+            "y": 184.0
+          },
+          "type": "basic"
+        },
+        "params": {
+          "alignment_column": null,
+          "alignment_table": null,
+          "databases": "[\"Uniref30_2302\", \"colabfold_envdb_202108\", \"PDB70_220313\"]",
+          "protein_column": null,
+          "protein_table": null,
+          "relaxed_prediction": false,
+          "use_templates": false
+        },
+        "status": "done",
+        "title": "Query OpenFold2"
+      },
+      "dragHandle": ".bg-primary",
+      "height": 653.0,
+      "id": "Query OpenFold2 1",
+      "position": {
+        "x": 750.0,
+        "y": 480.0
+      },
+      "type": "basic",
+      "width": 523.0
+    }
+  ]
+}

lynxkite-app/web/src/index.css

@@ -110,7 +110,7 @@ body {
     --status-color-1: oklch(75% 0.2 55);
     --status-color-2: oklch(90% 0.2 55);
     --status-color-3: oklch(75% 0.1 55);
-    /* animation: active-node-gradient-animation 2s ease-in-out infinite; */
+    animation: active-node-gradient-animation 2s ease-in-out infinite;
   }
 
   .lynxkite-node .title.planned {

lynxkite-bio/src/lynxkite_bio/__init__.py

@@ -1,68 +1,4 @@
 """An expansion for `lynxkite-graph-analytics` that provides algorithms for biological applications."""
 
-from lynxkite_graph_analytics import Bundle, RelationDefinition
-from lynxkite.core import ops
-import joblib
-import numpy as np
-import pandas as pd
-import rdkit.Chem
-import rdkit.Chem.rdFingerprintGenerator
-import rdkit.Chem.Fingerprints.ClusterMols
-import scipy
-
-mem = joblib.Memory(".joblib-cache")
-ENV = "LynxKite Graph Analytics"
-op = ops.op_registration(ENV)
-
-
-@op("Parse SMILES")
-def parse_smiles(bundle: Bundle, *, table="df", smiles_column="SMILES", save_as="mols"):
-    """Parse SMILES strings into RDKit molecules."""
-    df = bundle.dfs[table]
-    mols = [rdkit.Chem.MolFromSmiles(smiles) for smiles in df[smiles_column].dropna()]
-    mols = [mol for mol in mols if mol is not None]
-    bundle = bundle.copy()
-    bundle.dfs[table] = df.assign(**{save_as: mols})
-    return bundle
-
-
-def _get_similarity_matrix(mols):
-    mfpgen = rdkit.Chem.rdFingerprintGenerator.GetMorganGenerator(radius=2, fpSize=2048)
-    fps = [(0, mfpgen.GetFingerprint(mol)) for mol in mols]
-    similarity_matrix = rdkit.Chem.Fingerprints.ClusterMols.GetDistanceMatrix(
-        fps, metric=rdkit.Chem.DataStructs.TanimotoSimilarity, isSimilarity=1
-    )
-    return scipy.spatial.distance.squareform(similarity_matrix)
-
-
-@op("Graph from molecule similarity")
-def graph_from_similarity(
-    bundle: Bundle, *, table="df", mols_column="mols", average_degree=10
-):
-    """Creates edges for pairs of molecules that are the most similar."""
-    df = bundle.dfs[table]
-    mols = df[mols_column]
-    similarity_matrix = _get_similarity_matrix(mols)
-    i_idx, j_idx = np.triu_indices_from(similarity_matrix, k=1)
-    sim_values = similarity_matrix[i_idx, j_idx]
-    N = int(average_degree * len(mols))
-    top_n_idx = np.argsort(sim_values)[-N:]
-    top_n_pairs = [(i_idx[k], j_idx[k], sim_values[k]) for k in top_n_idx]
-    edges = pd.DataFrame(top_n_pairs, columns=["source", "target", "similarity"])
-    nodes = df.copy()
-    nodes.index.name = "id"
-    bundle = Bundle(
-        dfs={"edges": edges, "nodes": nodes},
-        relations=[
-            RelationDefinition(
-                df="edges",
-                source_column="source",
-                target_column="target",
-                source_table="nodes",
-                target_table="nodes",
-                source_key="id",
-                target_key="id",
-            )
-        ],
-    )
-    return bundle
+from . import nims  # noqa (imported to trigger registration)
+from . import rdkit  # noqa (imported to trigger registration)

lynxkite-bio/src/lynxkite_bio/nims.py

@@ -0,0 +1,133 @@
+"""Wrappers for BioNeMo NIMs."""
+
+from enum import Enum
+from lynxkite_graph_analytics import Bundle
+from lynxkite.core import ops
+import joblib
+import os
+
+NIM_URLS = os.environ.get("NIM_URLS", "http://localhost:8000").split(",")
+
+mem = joblib.Memory(".joblib-cache")
+ENV = "LynxKite Graph Analytics"
+op = ops.op_registration(ENV)
+
+
+class MSASearchTypes(Enum):
+    ALPHAFOLD2 = "ALPHAFOLD2"
+    ESM2 = "ESM2"
+
+
+class AlignmentFormats(Enum):
+    FASTA = "fasta"
+    A3M = "a3m"
+    STOCKHOLM = "stockholm"
+    CLUSTAL = "clustal"
+    PDB = "pdb"
+    PIR = "pir"
+    MSF = "msf"
+    TSV = "tsv"
+
+
+@op("MSA-search")
+@mem.cache
+def msa_search(
+    bundle: Bundle,
+    *,
+    protein_table: str,
+    protein_column: str,
+    e_value: float = 0.0001,
+    iterations: int = 1,
+    search_type: MSASearchTypes = MSASearchTypes.ALPHAFOLD2,
+    output_alignment_formats: list[AlignmentFormats] = [
+        AlignmentFormats.FASTA,
+        AlignmentFormats.A3M,
+    ],
+    databases: str = '["Uniref30_2302", "colabfold_envdb_202108", "PDB70_220313"]',
+):
+    bundle = bundle.copy()
+    return bundle
+
+
+@op("Query OpenFold2")
+@mem.cache
+def query_openfold2(
+    bundle: Bundle,
+    *,
+    protein_table: str,
+    protein_column: str,
+    alignment_table: str,
+    alignment_column: str,
+    use_templates: bool = False,
+    relaxed_prediction: bool = False,
+    databases: str = '["Uniref30_2302", "colabfold_envdb_202108", "PDB70_220313"]',
+):
+    bundle = bundle.copy()
+    return bundle
+
+
+@op("View molecules", view="visualization")
+def view_molecules(
+    bundle: Bundle,
+    *,
+    molecule_table: str,
+    molecule_column: str,
+    color="spectrum",
+):
+    return {
+        "series": [
+            {
+                "type": "pie",
+                "radius": ["40%", "70%"],
+                "itemStyle": {
+                    "borderRadius": 10,
+                    "borderColor": "#fff",
+                    "borderWidth": 2,
+                },
+                "data": [
+                    {"value": 2, "name": "Hydrogen"},
+                    {"value": 1, "name": "Sulfur"},
+                    {"value": 4, "name": "Oxygen"},
+                ],
+            }
+        ]
+    }
+
+
+@op("Known drug")
+def known_drug(*, drug_name: str):
+    return Bundle()
+
+
+@op("Query GenMol")
+@mem.cache
+def query_genmol(
+    bundle: Bundle,
+    *,
+    molecule_table: str,
+    molecule_column: str,
+    num_molecules: int = 5,
+    temperature: float = 1.0,
+    noise: float = 0.2,
+    step_size: int = 4,
+    scoring: str = "QED",
+):
+    bundle = bundle.copy()
+    return bundle
+
+
+@op("Query DiffDock")
+@mem.cache
+def query_diffdock(
+    proteins: Bundle,
+    ligands: Bundle,
+    *,
+    protein_table: str,
+    protein_column: str,
+    ligand_table: str,
+    ligand_column: str,
+    num_poses=10,
+    time_divisions=20,
+    num_steps=18,
+):
+    return proteins

lynxkite-bio/src/lynxkite_bio/rdkit.py

@@ -0,0 +1,68 @@
+"""An expansion for `lynxkite-graph-analytics` that provides algorithms for biological applications."""
+
+from lynxkite_graph_analytics import Bundle, RelationDefinition
+from lynxkite.core import ops
+import joblib
+import numpy as np
+import pandas as pd
+import rdkit.Chem
+import rdkit.Chem.rdFingerprintGenerator
+import rdkit.Chem.Fingerprints.ClusterMols
+import scipy
+
+mem = joblib.Memory(".joblib-cache")
+ENV = "LynxKite Graph Analytics"
+op = ops.op_registration(ENV)
+
+
+@op("Parse SMILES")
+def parse_smiles(bundle: Bundle, *, table="df", smiles_column="SMILES", save_as="mols"):
+    """Parse SMILES strings into RDKit molecules."""
+    df = bundle.dfs[table]
+    mols = [rdkit.Chem.MolFromSmiles(smiles) for smiles in df[smiles_column].dropna()]
+    mols = [mol for mol in mols if mol is not None]
+    bundle = bundle.copy()
+    bundle.dfs[table] = df.assign(**{save_as: mols})
+    return bundle
+
+
+def _get_similarity_matrix(mols):
+    mfpgen = rdkit.Chem.rdFingerprintGenerator.GetMorganGenerator(radius=2, fpSize=2048)
+    fps = [(0, mfpgen.GetFingerprint(mol)) for mol in mols]
+    similarity_matrix = rdkit.Chem.Fingerprints.ClusterMols.GetDistanceMatrix(
+        fps, metric=rdkit.Chem.DataStructs.TanimotoSimilarity, isSimilarity=1
+    )
+    return scipy.spatial.distance.squareform(similarity_matrix)
+
+
+@op("Graph from molecule similarity")
+def graph_from_similarity(
+    bundle: Bundle, *, table="df", mols_column="mols", average_degree=10
+):
+    """Creates edges for pairs of molecules that are the most similar."""
+    df = bundle.dfs[table]
+    mols = df[mols_column]
+    similarity_matrix = _get_similarity_matrix(mols)
+    i_idx, j_idx = np.triu_indices_from(similarity_matrix, k=1)
+    sim_values = similarity_matrix[i_idx, j_idx]
+    N = int(average_degree * len(mols))
+    top_n_idx = np.argsort(sim_values)[-N:]
+    top_n_pairs = [(i_idx[k], j_idx[k], sim_values[k]) for k in top_n_idx]
+    edges = pd.DataFrame(top_n_pairs, columns=["source", "target", "similarity"])
+    nodes = df.copy()
+    nodes.index.name = "id"
+    bundle = Bundle(
+        dfs={"edges": edges, "nodes": nodes},
+        relations=[
+            RelationDefinition(
+                df="edges",
+                source_column="source",
+                target_column="target",
+                source_table="nodes",
+                target_table="nodes",
+                source_key="id",
+                target_key="id",
+            )
+        ],
+    )
+    return bundle