PASTA for proteins.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate multiple sequence alignment (MSA) is crucial for phylogenetic analysis and understanding evolutionary relationships.
• Scaling MSA methods to large sequence datasets remains a significant computational challenge.
• Existing methods often face trade-offs between accuracy and computational efficiency.

Purpose of the Study:

• To enhance the scalability and accuracy of the PASTA multiple sequence alignment method.
• To evaluate the performance of PASTA with various underlying alignment algorithms on large datasets.
• To demonstrate improvements in alignment and phylogenetic tree accuracy using PASTA.

Main Methods:

• PASTA (Partitioning, Alignment, and Searching) employs a divide-and-conquer strategy with iteration.
• The study extended PASTA to incorporate multiple base alignment methods: MAFFT G-INS-i, MAFFT Homologs, CONTRAlign, and ProbCons.
• Performance was assessed using 224 diverse datasets from the BAliBASE 4 benchmark, focusing on datasets with at least 50 sequences.

Main Results:

• PASTA effectively enables accurate base alignment methods to scale to larger datasets.
• The enhanced PASTA demonstrated reduced computational effort for large-scale alignments.
• Significant improvements in alignment accuracy and phylogenetic tree reconstruction were observed on the largest BAliBASE datasets.

Conclusions:

• PASTA provides a scalable and accurate solution for multiple sequence alignment of large biological datasets.
• The integration of diverse base aligners enhances PASTA's flexibility and performance.
• PASTA represents a valuable tool for evolutionary and genomic research requiring high-accuracy alignments.