Compression-based classification of biological sequences and structures via the Universal Similarity Metric: experimental assessment

Table 4 SP-86-PDB

SP-86-PDB	UCD		NCD		CD
	UPGMA	NJ	UPGMA	NJ	UPGMA	NJ
Gzip	0.5372	0.5265	0.5450	0.5265	0.5372	0.5265
Bzip2	0.5411	0.5265	0.5400	0.5265	0.5440	0.5390
PPMd16	0.5367	0.5265	0.5468	0.5265	0.5477	0.5392
PPMd8	0.5346	0.5265	0.5384	0.5265	0.5477	0.5392
PPMd4	0.5346	0.5265	0.5371	0.5402	0.5388	0.5365
PPMd2	0.5377	0.5265	0.5528	0.5265	0.5439	0.5375
Huffman	0.5335	0.5265	0.5415	0.5265	0.5406	0.5265
Ac fast	0.5399	0.5265	0.5403	0.5265	0.5413	0.5265
Rc fast	0.5399	0.5389	0.5401	0.5299	0.5438	0.5265
Ac med.	0.5407	0.5355	0.5355	0.5276	0.5438	0.5265
Rc med.	0.5407	0.5265	0.5343	0.5317	0.5407	0.5265
Ac slow	0.5407	0.5291	0.5418	0.5265	0.5382	0.5265
Rc slow	0.5407	0.5332	0.5461	0.5283	0.5340	0.5265
BwtRleHuff	0.5376	0.5317	0.5429	0.5269	0.5487	0.5290
BwtMtfRleHuff	0.5374	0.5317	0.5409	0.5359	0.5487	0.5265
BwtRleAc fast	0.5386	0.5366	0.5593	0.5453	0.5481	0.5265
BwtMtfRleAc fast	0.5386	0.5265	0.5486	0.5302	0.5481	0.5265
BwtRleRc fast	0.5386	0.5265	0.5468	0.5265	0.5481	0.5265
BwtMtfRleRc fast	0.5364	0.5356	0.5478	0.5265	0.5481	0.5265
BwtRleRc med.	0.5376	0.5265	0.5439	0.5265	0.5481	0.5265
BwtMtfRleRc med.	0.5351	0.5265	0.5422	0.5265	0.5481	0.5265
BwtRleRc slow	0.5376	0.5265	0.5466	0.5265	0.5540	0.5302
BwtMtfRleRc slow	0.5376	0.5265	0.5426	0.5341	0.5442	0.5291
BwtWavelet	0.5376	0.5265	0.5376	0.5265	0.5449	0.5351

Experimental results for the SP-86-PDB data set, with the UCD (left), NCD (middle), and CD (right) distance. For each compression algorithm, we report the F-measure for both UPGMA and NJ methods.

ISSN: 1471-2105