-
Notifications
You must be signed in to change notification settings - Fork 7
kmtricks modules
kmtricks is composed of different modules which can be seen as building blocks of kmtricks pipeline.
These modules do not allow anything more than the pipeline but allow a finer processing (sub-sampling partitions, ect).
- kmtricks repart: Compute configuration and minimizer repartition
- kmtricks superk: Dispatch super-k-mers into partitions according to their minimizers.
- kmtricks count: Count k-mers/hashes in partitions.
- kmtricks merge: Merge equivalent partitions between samples.
- kmtricks format: Format Bloom filters matrices to obtains sample-specific Bloom filters.
Warning: kmtricks is highly tunable, some basic parameters, like k-mer size, are however maintained by a configuration file produced by kmtricks repart but for the sake of modularity many others are not fixed from the start. Some parameters or computations depend on the execution of previous modules with specific parameters (dependencies are described on each module page). This means that using these processing modules you are responsible for the consistency of the execution. Modules are only there for practical purposes for specific cases. You should only use them if you can't get exactly what you want using kmtricks pipeline.
- kmtricks dump: Dump kmtricks files in human readable format.
- kmtricks aggregate: Aggregate partition files.
Index modules (wrappers around a fork of HowDeSBT that works with kmtricks Bloom filters)
- kmtricks index: Build an HowDeSBT index using kmtricks Bloom filters.
- kmtricks query: Query an HowDeSBT index.