search

wwood / CoverM

Read coverage calculator for metagenomics
GNU General Public License v3.0
309 stars 31 forks source link

Strobealign aemb #225

Closed wwood closed 3 weeks ago

wwood commented 1 month ago

Thoughts @AroneyS ? Tests pass on my machine, something is wrong with the CI atm. It basically just runs strobealign --aemb, but can output dense or sparse, takes multiple refs and readsets etc, so a little more ergonomic.

Still needs to make a cached strobealign index if there are multiple readsets for a single reference. Also seems like it wouldn't be hard to adjust to genome mode too, just by taking a length-weighted average of the means it predicts.

Unfortunately it can't be used as a drop-in replacement for binning, because it only reports mean, not variance.

Once this is through, can copy the template so we can use fairy as well (apparently using it as an API isn't well supported).

ben

AroneyS commented 1 month ago

CI issue might be due to mambaforge sunsetting or mamba 2.0 (https://github.com/conda-incubator/setup-miniconda/issues/366). Bin Chicken's CI broke too, was fixed by removing mamba from the test workflow yml.

Also, cargo test errors locally for me:

  --- stderr
  thread 'main' panicked at /home/aroneys/.cargo/registry/src/index.crates.io-6f17d22bba15001f/bindgen-0.69.4/lib.rs:622:31:
  Unable to find libclang: "couldn't find any valid shared libraries matching: ['libclang.so', 'libclang-*.so', 'libclang.so.*', 'libclang-*.so.*'], set the `LIBCLANG_PATH` environment variable to a path where one of these files can be found (invalid: [])"
  note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
wwood commented 1 month ago

@AroneyS CI fixed - thanks for the tip. That looks like a strange linking error when you are running cargo test locally. Does the code make sense though?