What's next?
* External memory construction and memory-mapped queries, enabling Fulgor to work on datasets of any size.
* Distribution of ready-to-use Fulgor indices for large datasets.
(6/6)
With @jermp.bsky.social & @robp.bsky.social, we have just released Fulgor v4.1.0 (also on bioconda! bioconda.github.io/recipes/fulg...).
Here's what's new, and what's to come. 🧵 (1/6)
[Queries] We further improved queries by reducing redundant tasks and adding new output formats.
Besides ASCII, pseudoalignment results can now be written in binary and in a compressed format, leading to smaller output files and faster serialization. (2/6)
Interestingly, executing the same queries on the full index or on a subset (e.g. a single species) is equally fast, reinforcing the idea that a comprehensive index is better than a fragmented one. (5/6)