Fast, multithreaded pair-HMM counting of MaP-seq mutations.
Each release carries static cmuts binaries for Linux (x86_64, aarch64) and macOS (arm64). These bundle htslib and HDF5, so no libraries need to be installed to run them. Alternatively, you can build from source, which may provide marginal speedups.
cmuts align calls separate programs, which must be on the PATH.
On macOS with Homebrew:
brew install minimap2 samtools vsearchOn Debian and Ubuntu:
apt install minimap2 samtools vsearchcmuts plot requires a Python 3 installation with the h5py, numpy, and plotly packages.
python3 -m pip install h5py numpy plotlyCompute reactivity rates via the pair HMM on a specific experiment:
cmuts hmm -f references.fasta -o treated.h5 treated.bamSubtract rates computed from multiple experiments:
cmuts sub -o reactivity.h5 treated.h5 untreated.h5Normalize reactivity rates across experiments:
cmuts norm -o apo-normalized.h5 -o holo-normalized.h5 apo.h5 holo.h5Generate an interactive report of the results:
cmuts plot \
apo-normalized.h5 --label "Apo" \
holo-normalized.h5 --label "Holo"See the docs for more details on each program, the HDF5 outputs, and special use cases.
See CONTRIBUTING.md for information on development builds and running tests.