PhyClone 0.8.0

Breaking Changes

• --low-loss-prob CLI option has been removed
• Sample trace is now saved in the HDF5 binary file format, trace file extension should now be .h5
  • Trace files from previous PhyClone versions are incompatible with v0.8.0 and later

Features

• Sample trace is no longer saved as a python pickle file, now uses HDF5
• All output commands (map, consensus, and the topology archive of topology-report) can now additionally produce a sample prevalence table which reports the the clonal CCF and clonal prevalence estimates per sample
• The map and consensus commands have updated CLI options for the user to supply a path to save the new output table (using the -s / --out-sample-prev-table parameter)
• Array hashes have been updated from 64 to 128 bit
• Column requirements for the cluster file have been relaxed such that only mutation_id and cluster_id are minimally required (though the additional columns are still required for advanced outlier modelling options)

Performance Improvements

• Trace entries have been refactored to improve memory usage performance (in most cases)

Full Changelog: 0.7.1...0.8.0

PhyClone 0.7.1

What's Changed

• Input Validator Improvements by @elhurtado in #14

Bugfix:

• CLI no longer has redundant definition of '-t' parameter

Feature:

• Affected file path is now included in error message raised by InputValidator

Full Changelog: 0.7.0...0.7.1

PhyClone 0.7.0

What's Changed

• Speed improvements by @elhurtado in #12

Full Changelog: 0.6.0...0.7.0

PhyClone 0.6.0

What's Changed

• Input validation implemented by @elhurtado in #11
• Analysis runs without a user-supplied seed now report the machine-entropy based seed used

Full Changelog: v0.5.1...0.6.0

PhyClone 0.5.1

Introducing PhyClone: Accurate Bayesian reconstruction of cancer phylogenies from bulk sequencing.

This release version of PhyClone was used during benchmarking experiments.

What's Changed

• PhyClone method's first release 🥳
• Feature adding by @elhurtado in #1
• Current PhyClone by @elhurtado in #2
• Bringing in loss prior probability assignment feature by @elhurtado in #3
• Cleanup by @elhurtado in #4
• Topology report trees by @elhurtado in #5

New Contributors

• @elhurtado made their first contribution in #1
• Special thanks to @aroth85 for the initial software implementation, method conceptualization, and for being an excellent PI 🏅