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

Table 5 SP-86-ATOM

SP-86-ATOM	UCD		NCD		CD
	UPGMA	NJ	UPGMA	NJ	UPGMA	NJ
Gzip	0.5349	0.5337	0.5376	0.5328	0.5338	0.5396
Bzip2	0.5779	0.5510	0.5779	0.5510	0.5632	0.5472
PPMd16	0.5425	0.5460	0.5361	0.5265	0.5567	0.5265
PPMd8	0.5550	0.5550	0.5443	0.5569	0.5562	0.5550
PPMd4	0.5454	0.5459	0.5394	0.5571	0.5482	0.5472
PPMd2	0.5348	0.5297	0.5412	0.5265	0.5418	0.5278
Huffman	0.5303	0.5265	0.5303	0.5265	0.5331	0.5365
Ac fast	0.5553	0.5385	0.5625	0.5524	0.5645	0.5413
Rc fast	0.5587	0.5477	0.5626	0.5389	0.5602	0.5472
Ac med.	0.5580	0.5493	0.5563	0.5438	0.5627	0.5272
Rc med.	0.5581	0.5583	0.5510	0.5434	0.5534	0.5492
Ac slow	0.5440	0.5314	0.5410	0.5265	0.5471	0.5463
Rc slow	0.5363	0.5265	0.5376	0.5265	0.5489	0.5328
BwtRleHuff	0.5408	0.5390	0.5408	0.5332	0.5557	0.5509
BwtMtfRleHuff	0.5411	0.5438	0.5411	0.5438	0.5420	0.5567
BwtRleAc fast	0.5365	0.5282	0.5365	0.5282	0.5323	0.5363
BwtMtfRleAc fast	0.5775	0.5421	0.5775	0.5421	0.5558	0.5747
BwtRleRc fast	0.5317	0.5362	0.5365	0.5462	0.5397	0.5265
BwtMtfRleRc fast	0.5791	0.5421	0.5791	0.5609	0.5558	0.5550
BwtRleRc med.	0.5338	0.5265	0.5338	0.5284	0.5340	0.5317
BwtMtfRleRc med.	0.5390	0.5550	0.5390	0.5550	0.5495	0.5405
BwtRleRc slow	0.5350	0.5385	0.5385	0.5419	0.5415	0.5415
BwtMtfRleRc slow	0.5338	0.5354	0.5338	0.5354	0.5420	0.5694
BwtWavelet	0.5362	0.5344	0.5362	0.5368	0.5339	0.5265

Experimental results for the SP-86-ATOM 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