Methods
Alignment-free methods implemented in Alfree
Methods based on k-mers words frequencies
| Distance | Method name | Seq | Word size | Vector type | References |
|---|---|---|---|---|---|
| d E | Squared Euclidean distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 1, 2 |
| d S | Euclidean distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 3 |
| d Eseq1 | Squared Euclidean distance normalized by sequence length | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 4, 5 |
| d Eseq2 | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | ||
| d EVOL1 | Angle cosine evolutionary distance 1 | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 6, 7, 8 |
| dEVOL2 | Angle cosine evolutionary distance 2 | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 9 |
| d CV | Composition distance | nt/aa | 3rd order Markov model (≥3) | c | 9, 10 |
| d 2 | d2 distance | nt/aa | ≥2 | c, f, cw, fw, fstd, fstdw | 3, 11, 12 |
| d Minkowski | Minkowski distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 13 |
| d Manhattan | Manhattan (City Block) distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 13 |
| d abs_mean | Absolute difference among k-word vectors | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 4, 5 |
| d abs_mult | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | ||
| d abs_mult1 | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | ||
| d abs_mult2 | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | ||
| d Bray-Curtis | Bray-Curtis distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 14 |
| d Canberra | Canberra distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 14 |
| d Chebyshev | Chebyshev distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 14 |
| d KL | Kullback-Leibler discrepancy | nt/aa | ≥1 | f | 15 |
| d google | Normalized Google Distance | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 16 |
| d LCC | linear correlation coefficient | nt/aa | ≥1 | c, f, cw, fw, fstd, fstdw | 17 |
| d W | W-metric | aa | 1 | c | 18 |
| d FCGR | Frequency Chaos Game Representation | nt | ≥1 | FCGR-specific vector | 19, 20 |
| d RTD | Return Time Distribution | nt/aa | ≥1 | RTD-specific vector | 21 |
| d FFP | Feature Frequency Profiles | nt/aa | ≥1 | f | 22 |
| d Sorensen-Dice | Sørensen–Dice coefficient (Czekanowski's binary index) | nt/aa | ≥1 | nominal | 23 |
| d Jaccard | Jaccard index | nt/aa | ≥1 | nominal | 23 |
| d Hamming | Hamming distance | nt/aa | ≥1 | nominal | 24 |