In more distant species, DNA-based chains from part G are not sufficiently sensitive, but it is easier to find similarity between proteins
In this type of track you extract protein sequences based on genome sequence and gene annotation, then you compare protein sets from the two species and map protein alignments back to the genome
Commands from the last year create a psl file and load it. Then the alignments cannot be used to move between genomes. It would be better to convert psl to chain as in parts F and G.