Giulio Ermanno Pibiri

Currently, `boost` is used: - for the preprocessor's `for_each`; - for memory mapped files; - for iterating through gzipped files.

Create one master tool `tongrams` with sub-tools, like `tongrams build` and `tongrams query` instead of many separate executables.

**New topics** - Binomial Code (i.e., RRR) - Simple Dense Coding (SDC) - Front-Coding - Burrows-Wheeler Transform (BWT) and BZip - Predictions by Partial Matching (PPM) - Range Coding -...

As per the title, try to avoid that. So annoying. Perhaps a simple linear search over the buckets' thresholds is fast anyway since these are supposed to be very few.

The output is described here: https://github.com/COMBINE-lab/cuttlefish#cuttlefish-1-output. We want to parse the `.cf_seg` files output by Cuttlefish1 to build the unitigs from a reference dBG.

Store the minimizer's absolute positions rather than those of super-kmers. Align the query kmer to the minimizer's position at query time. This will simplify the code logic avoiding the scan...

See https://github.com/jermp/essentials/blob/master/include/essentials.hpp#L121.

Hi @dnbaker and thank you for this great library! Consider this little fix that avoids warnings on both Linux and Mac. Best, -Giulio

different runs return different number of unitigs and distinct color subsets

6

Hello @Guilucand and @alexandrutomescu. Let me first congratulate with you for this excellent algorithm! We are currently using GGCAT for building [Fulgor](https://github.com/jermp/fulgor), a colored k-mer index based on SSHash (same...