Faithful to the paper
The method of Tharanga et al. (2025), with every formula and source documented, validated on the paper’s own HIV-1 and MERS-CoV datasets.
This is the documentation of DiMA . The newest release is .
This documentation of the development version includes changes that are not released yet. The newest release is .
For every overlapping k-mer position of a viral protein or gene alignment, DiMA counts the k-mers and the sequences that support them, estimates their Shannon entropy with the paper’s correction for sample size, classifies variants as Index, Major, Minor and Unique motifs with their incidences, can count them per host, country, year or any header field, and finds the historically conserved sequences of the alignment (Tharanga et al. 2025).
Faithful to the paper
The method of Tharanga et al. (2025), with every formula and source documented, validated on the paper’s own HIV-1 and MERS-CoV datasets.
Reproducible
No random numbers: identical results on every platform and thread count, with the full provenance of each run recorded next to the results.
Fast and frugal
One streaming pass over plain or compressed alignments; memory grows with diversity, not with the number of sequences.
Command line and desktop
dima for scripts and clusters, the desktop app for exploring entropy, motifs and conserved
regions interactively — one library behind both.
If you use DiMA, please cite the paper and the version you used; dima cite prints both, and
dima cite --from results.json the references of a run. See Citing DiMA.