liteseq

liteseq is a lightweight C library for parsing GFA files, using a bidirected model of the graph. In this model, each vertex has two sides (or ends): a left side and a right side.

Features and Limitations

• Configurable Parsing: Choose whether to include vertex labels, references, etc.
• Minimal Graph: Improve performance by fetching only what's needed
• GFA Version Support:
  • GFA v1.0 (VN:Z:1.0)
• Limitations
• Sequence IDs in the GFA should be numerical

Configuration

Often, we only need to read a subset of the GFA data. Therefore, liteseq depends on a config to determine what graph properties to read from the GFA. This can be seen in examples/main.c.

When both inc_vtx_labels and inc_refs are false it only reads the most minimal graph possible, that is, vertex IDs and the edges.

Building liteseq

Prerequisites:

• CMake (3.0+ recommended)
• C compiler (e.g., GCC or Clang)

Fetch the source code

git clone https://github.com/urbanslug/liteseq.git

cd liteseq

Release mode

cmake -DCMAKE_BUILD_TYPE=Release  -H. -Bbuild && cmake --build build -- -j 3

Debug mode

cmake -DCMAKE_BUILD_TYPE=Debug  -H. -Bbuild && cmake --build build -- -j 3

Building a Specific Target

If you want to compile a specific target (such as examples/main.c), run:

cmake -H. -DCMAKE_BUILD_TYPE=Debug -Bbuild && cmake --build build --target liteseq -- -j 8

Usage and Examples

1. Include Headers:

#include <gfa.h>

2. Set Up Configuration:

gfa_config config = {
    .inc_vtx_labels = false,
    .inc_refs = false,
    // ...
};

3. Initialize, Parse, and Use:

gfa_props *gfa = gfa_new(config);
// ...
// parse, process, etc.

4. Cleanup:

gfa_free(gfa);

See examples/main.c for more detailed usage.

Contributing

Contributions, bug reports, and feature requests are welcome. Please open an issue or a pull request on GitHub.

License

MIT

Happy parsing!