StrobeMap (C++) v0.0.2

Many updates to the C++ implementation of StrobeMap

1. NAM finding is now parallelized (OpenMP).
2. StrobeMap can take read query sequences both as fasta and fastq.
3. Faster indexing of references (through preallocation of hash table).
4. Fixed minor bugs such as seg fault for references where no strobemer could be generated (e.g., reference containing only N's)
5. Precompiled binaries added for Linux and Mac OSx.
6. Several parameters added, described below.

• Parameter -s to split the output into one file per thread and fw/rc NAMs (to be compatible as seed finder for uLTRA RNA long-read aligner)
• Parameter -L to print at most L NAMs per query [1000]. Will print the NAMs with highest score S = n_strobemer_hits * query_span.
• Parameter -S to sort output NAMs for each query based on score. Default is to sort first by ref ID, then by query coordinate, then by reference coordinate.

StrobeMap v0.0.1

The first stable working version of StrobeMap (C++ implementation). Implements randstrobes and hybridstrobes of orders 2 and 3, and minstrobes of order 2. Uses bitpacking for fast construction time. Computes NAMs between a set of queries mapped to a set of references.

TODO:

• Read fastq
• Parallelize mapping
• Separate creation and storage of reference index
• Option to output the NAM-length on both the query and the reference (5 column format)
• [Implement thinning?]