escpod subset¶

Split reads from a POD5 file into multiple output files based on a CSV mapping.

Usage¶

escpod subset --csv <FILE> -o <OUTPUT_DIR> [OPTIONS] <INPUT>

Arguments¶

Options¶

CSV File Format¶

The CSV file must have a header row with read_id and output columns:

read_id,output
a1b2c3d4-e5f6-7890-abcd-ef1234567890,sample1.pod5
b2c3d4e5-f6a7-8901-bcde-f12345678901,sample1.pod5
c3d4e5f6-a7b8-9012-cdef-123456789012,sample2.pod5

Supported UUID Formats¶

• Standard format: a1b2c3d4-e5f6-7890-abcd-ef1234567890
• No dashes: a1b2c3d4e5f67890abcdef1234567890
• Uppercase: A1B2C3D4-E5F6-7890-ABCD-EF1234567890

Notes¶

• Whitespace is trimmed from fields
• Empty lines are ignored
• Each unique output value creates a separate POD5 file

Examples¶

Basic Subsetting¶

Create a mapping file:

cat > mapping.csv << EOF
read_id,output
a1b2c3d4-e5f6-7890-abcd-ef1234567890,barcode01.pod5
b2c3d4e5-f6a7-8901-bcde-f12345678901,barcode01.pod5
c3d4e5f6-a7b8-9012-cdef-123456789012,barcode02.pod5
EOF

Split the POD5 file:

escpod subset experiment.pod5 --csv mapping.csv -o demultiplexed/

Demultiplexing by Barcode¶

If you have demultiplexing results from basecalling:

# Assuming you have a CSV from your demultiplexing pipeline
escpod subset multiplexed.pod5 --csv barcode_assignments.csv -o demux/

Splitting by Sample¶

Split reads into per-sample files:

escpod subset pooled.pod5 --csv sample_mapping.csv -o samples/

Output¶

The command prints a summary of the subsetting operation:

Subsetting 10000 reads into 3 output file(s)
Subsetting [████████████████████████████████████████] 10000/10000

Subset summary:
  Matched reads: 9500
  Unmatched reads: 500

Output files:
  demux/barcode01.pod5 (4000 reads)
  demux/barcode02.pod5 (3500 reads)
  demux/barcode03.pod5 (2000 reads)

Notes¶

• Reads not in the CSV mapping are excluded from output
• Run info is copied to all output files
• Each output file is a complete, valid POD5 file
• Signal data is decompressed and re-compressed for each output