Dolazactree

Function

dolazactree(
    self,
    tool_cna_files: List[str],
    tool_names: List[str],
    outfile: Optional[str] = "parsimony_score.csv"
) -> pd.DataFrame

This function evaluates phylogenetic parsimony scores for inferred CNA profiles using the external tool LAZAC.

For each tool:

1. Converts the CNA matrix into LAZAC-compatible input format.
2. Runs LAZAC with nearest-neighbor interchange (NNI) tree search.
3. Extracts the final parsimony score from the LAZAC output.
4. Collects scores across tools into a summary table.

The function writes a single CSV file containing the parsimony scores of all tools.

---

Parameters

Name	Type	Description
tool_cna_files	List[str]	List of CNA profile CSV files from different tools. Must align with tool_names in order.
tool_names	List[str]	Tool names used for labeling outputs and result table.
outfile	Optional[str]	Output filename for the summary table. Default: "parsimony_score.csv".

---

Input File Format

Each file in tool_cna_files is expected to be a CNA matrix:

• rows: genomic regions
• columns: cell IDs
• values: CNA states (e.g., "1|1", "2|1")

Example:

region,cell_001,cell_002,cell_003
chr1:1-100000,1|1,1|1,2|1
chr1:100001-200000,1|1,1|1,2|1

Before running LAZAC:

• The CNA matrix is converted into a LAZAC-compatible input file via create_lazac_input.
• The converted file is saved as:

{output_dir}/lazac_{tool_name}/{tool_name}_cn_profile.csv

---

External Dependency

This function requires:

• lazac executable available in the system PATH.
• The command executed is:

lazac nni <cn_profile_file> -a 2 -o <output_prefix>

Where:

• nni performs nearest-neighbor interchange search.
• -a 2 specifies diploid assumption.
• -o sets output prefix.

---

Output

For each tool, a dedicated directory is created:

{self.output_dir}/lazac_{tool_name}/

LAZAC generates output files including:

• {tool_name}_info.json (contains final parsimony score)
• additional tree-related outputs

Summary Table (outfile)

Saved to:

os.path.join(self.output_dir, outfile)

Structure:

Column	Meaning
Tool	Tool name
Parsimony	Final parsimony score extracted from LAZAC output

Example:

Tool,Parsimony
CHISEL,14331085
Alleloscope,34
SIGNALS,5577962

Lower parsimony scores indicate simpler evolutionary histories under the parsimony model.

---

Return Value

Returns a pd.DataFrame containing:

Column	Meaning
Tool	Tool name
Parsimony	Parsimony score

---

Example

from hcbench.realbench.realbench import RealBench

bench = RealBench(output_dir="out/real_output")

bench.dolazactree(
    tool_cna_files=[
        "/path/to/chisel/cna.csv",
        "/path/to/signals/cna.csv",
    ],
    tool_names=["CHISEL", "SIGNALS"],
    outfile="parsimony_score.csv"
)

---