castanetgraph

castanetgraph.py processes Castanet per-probe depth/call output together with precomputed graph-path data, then reports top-supported graph paths and optional depth/consensus artifacts.

Usage

python castanetgraph.py \
  --inputgraphdata /path/to/graphdata.parquet \
  --castanetfolder /path/to/castanet_sample_folder \
  --output /path/to/output/sample_prefix

Required arguments

• --inputgraphdata: Parquet file containing graph/path summaries.
• --castanetfolder: Castanet output folder containing *_depth.csv plus optional depth/consensus subfolders.
• --output: Output prefix used for all generated files.

Expected Castanet folder layout

At minimum:

• <castanetfolder>/*_depth.csv

Optional (used for plots/consensus sequence export if present):

• <castanetfolder>/Depth_output/*depth_by_pos.csv
• <castanetfolder>/consensus_data/<probe_or_block_id>/*_remapped_consensus_sequence.fasta

Main outputs

Using --output /path/prefix, the script writes:

• /path/prefix_top_paths.tsv (created only when no paths remain after filtering)
• /path/prefix_all_targets_w_top_paths.tsv (main merged call table)
• /path/prefix_rMLST_combined.tsv (when rMLST grouping is detected)
• /path/prefix_depthplots/ (depth plots when depth files exist)
• /path/prefix_consensus/ (consensus FASTA files when consensus input exists)