What are the differences between BLAST and BLAT?
BLAST (Basic Local Alignment Search Tool) finds regions of local similarity among nucleotide or amino acid sequences. It compares a query sequence (DNA or protein) to a large set of sequences (the target) and calculates the statistical significance of matches. From Ensembl release 71 onwards, Ensembl uses the NCBI Blast implementation for its search options. Use Ensembl BLAST or read our tool help. More information can be found at NCBI.
BLAT (BLAST-like Alignment Tool) is a sequence alignment tool similar to BLAST but structured differently. BLAT quickly finds similarity in DNA and protein but it needs an exact or nearly-exact match to find a hit. Therefore Blat is not as flexible as BLAST. Since BLAST can find much more remote matches than Blat, it is the recommended tool when searching more distantly related sequences.
Due to its faster speed, BLAT is the default similarity search program in the Ensembl page for both nucleotide query and nucleotide target sequences. For more information about BLAT, click here.
Patches and haplotypes
If you are looking for sequence that may be on a patch or haplotype, use BLAST, not BLAT. BLAT indices are built using the primary_assembly.fa file, whereas Blast indices are built from the toplevel.fa files and therefore will provide hits to both the primary assembly and the patches. You can use a patch/haplotype sequence as a query sequence to find where it maps on the genome, but you need to choose BLAST instead of BLAT to allow for mismatches.