amina-cli 0.2.1__tar.gz → 0.2.2__tar.gz

{amina_cli-0.2.1 → amina_cli-0.2.2}/.gitignore

@@ -39,8 +39,18 @@ uv.lock
 
 # OS
 .DS_Store
+.DS_Store?
+._*
+.AppleDouble
+.LSOverride
+.Spotlight-V100
+.Trashes
+Icon?
 Thumbs.db
 
+# PyMOL
+*.pml
+
 # Testing
 .pytest_cache/
 .coverage

{amina_cli-0.2.1 → amina_cli-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: amina-cli
-Version: 0.2.1
+Version: 0.2.2
 Summary: CLI for AminoAnalytica protein engineering platform
 Project-URL: Homepage, https://aminoanalytica.com
 Project-URL: Documentation, https://docs.aminoanalytica.com

{amina_cli-0.2.1 → amina_cli-0.2.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "amina-cli"
-version = "0.2.1"
+version = "0.2.2"
 description = "CLI for AminoAnalytica protein engineering platform"
 readme = "README.md"
 license = {text = "Apache-2.0"}

{amina_cli-0.2.1 → amina_cli-0.2.2}/src/amina_cli/__init__.py

@@ -9,4 +9,4 @@ Quick start:
     amina run esmfold --sequence "MKFLILLFNILCLFPVLAADNH"
 """
 
-__version__ = "0.2.1"
+__version__ = "0.2.2"

amina_cli-0.2.2/src/amina_cli/commands/tools/properties/esm1v.py

@@ -0,0 +1,175 @@
+"""ESM-1v variant effect prediction tool for the Amina CLI."""
+
+import typer
+from pathlib import Path
+from typing import Optional
+from rich.console import Console
+
+METADATA = {
+    "name": "esm1v",
+    "display_name": "ESM-1v Variant Effect",
+    "category": "properties",
+    "description": "Predict variant effects on protein function using ESM-1v masked marginal scoring",
+    "modal_function_name": "esm1v_worker",
+    "modal_app_name": "esm1v-api",
+    "status": "available",
+    "outputs": {
+        "scores_csv_filepath": "Mutation scores CSV",
+        "heatmap_filepath": "DMS heatmap (PNG, --mode dms only)",
+        "summary_json_filepath": "Summary statistics (JSON)",
+    },
+}
+
+console = Console()
+
+
+def register(app: typer.Typer):
+    """Register this tool's command with the app."""
+    from amina_cli.commands.tools import run_tool_with_progress
+
+    @app.command("esm1v")
+    def run_esm1v(
+        mode: str = typer.Option(
+            ...,
+            "--mode",
+            "-m",
+            help="Scoring mode: single, batch, or dms (deep mutational scan)",
+        ),
+        sequence: Optional[str] = typer.Option(
+            None,
+            "--sequence",
+            "-s",
+            help="Protein sequence (amino acid letters, max 1022 residues)",
+        ),
+        fasta: Optional[Path] = typer.Option(
+            None,
+            "--fasta",
+            "-f",
+            help="Path to FASTA file containing the protein sequence",
+            exists=True,
+        ),
+        mutations: Optional[str] = typer.Option(
+            None,
+            "--mutations",
+            help="Comma-separated mutations in XposY format, e.g. 'A56L,D78G'",
+        ),
+        mutations_csv: Optional[Path] = typer.Option(
+            None,
+            "--mutations-csv",
+            help="Path to CSV file with mutations (column: mutation)",
+            exists=True,
+        ),
+        num_models: int = typer.Option(
+            5,
+            "--num-models",
+            "-n",
+            help="Number of ESM-1v ensemble checkpoints (1-5)",
+        ),
+        output: Optional[Path] = typer.Option(
+            None,
+            "--output",
+            "-o",
+            help="Output directory for results (required unless --background)",
+        ),
+        job_name: Optional[str] = typer.Option(
+            None,
+            "--job-name",
+            "-j",
+            help="Custom job name for output files (default: random 4-letter code)",
+        ),
+        background: bool = typer.Option(
+            False,
+            "--background",
+            help="Submit job and return immediately without waiting for completion",
+        ),
+    ):
+        """
+        Predict variant effects on protein function using ESM-1v masked marginal scoring.
+
+        Scores how amino acid substitutions affect protein fitness using masked
+        marginal scoring with up to 5 ensemble checkpoints.
+
+        Modes:
+          - single: Score one mutation (e.g., --mutations "A56L")
+          - batch:  Score multiple mutations (comma-separated or CSV)
+          - dms:    Full deep mutational scan (all 20 AAs at every position)
+
+        Examples:
+            amina run esm1v -m single -s "MKFLILLFNILCLFPVLAADNH" --mutations "M1A" -o ./output/
+            amina run esm1v -m batch -f seq.fasta --mutations "M1A,K2R,L5W" -o ./output/
+            amina run esm1v -m batch -f seq.fasta --mutations-csv muts.csv -o ./output/
+            amina run esm1v -m dms -s "MKFLILLFNILCLFPVLAADNH" -o ./output/
+            amina run esm1v -m dms -f seq.fasta -n 3 -o ./output/  # 3-model ensemble
+        """
+        # Validate required options
+        if output is None and not background:
+            console.print("[red]Error:[/red] --output / -o is required (unless using --background)")
+            raise typer.Exit(1)
+
+        # Validate mode
+        valid_modes = ["single", "batch", "dms"]
+        if mode.lower() not in valid_modes:
+            console.print(f"[red]Error:[/red] Invalid mode '{mode}'. Choose from: {', '.join(valid_modes)}")
+            raise typer.Exit(1)
+        mode = mode.lower()
+
+        # Validate num_models
+        if num_models < 1 or num_models > 5:
+            console.print("[red]Error:[/red] --num-models must be between 1 and 5")
+            raise typer.Exit(1)
+
+        # Validate sequence input (exactly one)
+        if sequence is None and fasta is None:
+            console.print("[red]Error:[/red] Provide --sequence or --fasta")
+            raise typer.Exit(1)
+        if sequence is not None and fasta is not None:
+            console.print("[red]Error:[/red] Provide only one of --sequence or --fasta")
+            raise typer.Exit(1)
+
+        # Validate mutation input based on mode
+        if mode in ("single", "batch"):
+            if mutations is None and mutations_csv is None:
+                console.print(f"[red]Error:[/red] Mode '{mode}' requires --mutations or --mutations-csv")
+                raise typer.Exit(1)
+            if mutations is not None and mutations_csv is not None:
+                console.print("[red]Error:[/red] Provide only one of --mutations or --mutations-csv")
+                raise typer.Exit(1)
+        elif mode == "dms":
+            if mutations is not None or mutations_csv is not None:
+                console.print("[red]Error:[/red] Mode 'dms' does not accept mutations input")
+                raise typer.Exit(1)
+
+        # Build params
+        params = {
+            "mode": mode,
+            "num_models": num_models,
+        }
+
+        # Sequence input
+        if sequence:
+            params["sequence"] = sequence.upper()
+            console.print(f"Sequence: {len(sequence)} residues")
+        elif fasta:
+            fasta_content = fasta.read_text()
+            if not fasta_content.strip().startswith(">"):
+                console.print("[red]Error:[/red] File does not appear to be in FASTA format")
+                raise typer.Exit(1)
+            params["fasta_content"] = fasta_content
+            console.print(f"Read sequence from {fasta}")
+
+        # Mutation input
+        if mutations:
+            params["mutations"] = mutations
+            num_muts = len([m for m in mutations.split(",") if m.strip()])
+            console.print(f"Mutations: {num_muts} specified")
+        elif mutations_csv:
+            params["mutations_csv_content"] = mutations_csv.read_text()
+            console.print(f"Read mutations from {mutations_csv}")
+
+        if job_name:
+            params["job_name"] = job_name
+
+        console.print(f"Mode: {mode} | Ensemble: {num_models} model(s)")
+
+        # Execute
+        run_tool_with_progress(METADATA["name"], params, output, background=background)