BAliBASE

[…] work to determine the optimum parameters. The current parameters pertaining to the objective function in most literature are empirical values which are independently associated with the sequences []. BALiBASE is a database of manually refined multiple sequence alignments [] and is usually used to test performance of MSA method [].Many open source online alignment tools are available that can align […]

[…] In this work the universally known benchmark datasets such as Benchmark Alignment Database (BAliBASE 3.0), Prefab 4.0, SABmark 1.65 and Oxbench 1.3 is used to examine and compare with various multiple sequence alignment methods. The BAliBASE 3.0 database contains 6255 protein sequences in to […]

[…] nnet160 matrix for partition function calculation by VTML200, which was proven to be more accurate. This was followed by training partition function parameters, i.e., gap penalties and temperature on BAliBASE 3 benchmark with a use of NOMAD algorithm for optimization of non-smooth functions. Another changes were introduced in order to shorten execution time. They include redesigning graphics proce […]

[…] algorithms such as MSAProbs, Probalign, MAFFT, Procons, Clustal Omega, T-Coffee, Kalign, MUSCLE, FSA, DIALIGN, PRANK, and CLUSTALW. shows the average TC scores of these algorithms on six subsets of BAliBASE 3.0. The data used in are drawn from Ref. , except the data about IBBOMSA. The proposed IBBOMSA is the fourth best aligner in terms of accuracy. The top aligners are MSAProbs, which reach th […]

[…] or ease of reference, we use GLProbs-UPGMA to refer the original GLProbs, and GLProb-WPGMA to refer the modified one. We used both tools to align the normally related families in SABmark, OXBench and BALiBASE, and compute the TC scores of the resulting alignments, which is one of the most commonly used performance measure for evaluating multiple sequence alignments; the higher the scores, the bett […]

[…] To test the performance of the program, we used the most recent test set (Reference 10) in the BAliBASE benchmark suite [], composed of 218 reference alignments and containing a total of 17,892 protein sequences. These reference alignments are designed to reflect some of the problems specific t […]