fuse-ranges

name	fuse-ranges
description	Use when you need to merge overlapping or adjacent strand-specific alignment ranges encoded as comma-separated `start..end` lists in EDirect tables.
disable-model-invocation	true
user-invocable	true

fuse-ranges

CLI tool from the Entrez Direct (EDirect) package for fusing overlapping or adjacent ranges.

Quick Start

• Command: printf '5\t1\t+\t3..8,8..10,15..13\n' | PATH=/home/vimalinx/miniforge3/envs/bio/bin:$PATH /home/vimalinx/miniforge3/envs/bio/bin/fuse-ranges
• Local executable: /home/vimalinx/miniforge3/envs/bio/bin/fuse-ranges
• Reference: See references/help.md for full documentation

When To Use This Tool

• Expand comma-separated start..end range lists and fuse overlaps per strand.
• Collapse adjacent intervals too, not just strict overlaps.
• Turn EDirect-style alignment summaries into strand / start / end / length blocks for later filtering or reporting.

Common Patterns

# 1) Fuse a single row containing multiple strand-specific ranges
printf '5\t1\t+\t3..8,8..10,15..13\n' | \
  PATH=/home/vimalinx/miniforge3/envs/bio/bin:$PATH \
  /home/vimalinx/miniforge3/envs/bio/bin/fuse-ranges

+   3   10  8
+   13  15  3

# 2) Use in a larger EDirect table pipeline
some_edirect_command | \
  PATH=/home/vimalinx/miniforge3/envs/bio/bin:$PATH \
  /home/vimalinx/miniforge3/envs/bio/bin/fuse-ranges \
  > fused.tsv

Recommended Workflow

1. Generate a four-column table where column 3 is strand and column 4 is a comma-separated range list like 3..8,8..10,15..13.
2. Put the EDirect bin directory on PATH so sort-table is available to the wrapper.
3. Pipe the table through fuse-ranges and capture the merged strand / start / end / length output.
4. Spot-check that adjacent intervals were merged the way you expect.

Guardrails

• This wrapper depends on sibling EDirect tools such as sort-table; calling the absolute path alone can still fail with command not found.
• It prefilters with grep '^[1-9]', so rows starting with 0, whitespace, or non-digits are silently dropped.
• After that filter, only the strand column and the comma-separated range-list column are used; the first two fields are effectively ignored.
• Empty or non-matching input can yield a bogus 0 0 1 row instead of a clean error, so validate outputs before using them downstream.