String Partition for Building Big BWTs

The Burrows-Wheeler transform (BWT) is a reversible text transformation used extensively in text compression, indexing, and bioinformatics, particularly in the alignment of short reads. However, constructing the BWT for long strings poses significant challenges. We introduce a novel approach to partition a long string into shorter substrings, enabling the use of multi-string BWT construction algorithms to process these inputs. The approach partitions based on a prefix of the suffix array and we provide an implementation for DNA sequences. Through comparison with state-of-the-art BWT construction algorithms, we demonstrate a speed improvement of approximately 12% on a real genome dataset consisting of 3.2 billion characters. The proposed partitioning strategy is applicable to strings of any alphabet.