Bwa single end reads
A probabilistic short read aligner based on the use of position specific scoring matrices (PSSM).The aligner is adaptable in the sense that it can take into account the quality scores of the reads and models of data specific biases, such as those observed in Ancient DNA, PAR-CLIP data or genomes with biased nucleotide compositions.It is made for resequencing projects, namely in a diagnostic setting. GMAP: longer reads, with multiple indels and splices (see entry above under Genomics analysis); GSNAP: shorter reads, with one indel or up to two splices per read.
Computes Smith-Waterman gapped alignments and mapping qualities on one or more GPUs. Processes 100,000 to 500,000 reads per second (varies with data, hardware, and configured sensitivity).Performs affine-transform-optimized global alignment, which is slower but more accurate than Smith-Waterman. Can handle insertions, deletions, SNPs, and color errors (can map ABI SOLi D color space reads). Runs the Burrows-Wheeler Aligner-BWA on a Hadoop cluster.Handles Illumina, 454, Pac Bio, Sanger, and Ion Torrent data. Explicit time and accuracy tradeoff with a prior accuracy estimation, supported by indexing the reference sequences. It supports the algorithms BWA-MEM, BWA-ALN, and BWA-SW, working with paired and single reads.BWA-MEM and BWA-SW share similar features such as long-read support and split alignment, but BWA-MEM, which is the latest, is generally recommended for high-quality queries as it is faster and more accurate. But I am afraid the output could include something you do not necessarily want to (since there is no specific flag for the split reads).Maybe a better solution: you could switch to some mapper that will report the proper split in the hit of the read map, such as minimap2 from the same author.