Exact indexes accelerate exact and seeded search, but read mapping still needs approximate matching and biological interpretation.

highlighted = computed this step

Exact matching is only the starting point

Suffix arrays and the BWT find exact matches very fast. Real reads carry errors and real genomes carry variants, so read mapping needs approximate matching with mismatches, insertions, and deletions. The index accelerates that search; it does not solve it alone.

exact indexcomplete read aligner\text{exact index} \ne \text{complete read aligner}
Exact indexing boundaryThe table separates the exact data structure from biological interpretation.Exact index answersStill neededwhere this string occursapproximate read mappingexact deterministic matcheserrors and variantslocations in a stringbiological meaning

Production indexes add engineering

Here the data structures are exact and deterministic on a small string. Whole-genome indexes add compression and engineering for scale, and they answer where an exact string occurs, not what the sequence means.

where it occurswhat it means\text{where it occurs} \ne \text{what it means}
Exact indexing boundaryThe table separates the exact data structure from biological interpretation.Exact index answersStill neededwhere this string occursapproximate read mappingexact deterministic matcheserrors and variantslocations in a stringbiological meaning