Files
obitools4/pkg/obikmer/kmer_distance_test.go
T
Eric Coissac a0d8187fd6 Implement k-mer distance metrics and constrained generation
Adds support for Hamming distance, constrained k-mer generation, and optimized Longest Common Subsequence (LCS) calculations using scalar, batch, and SIMD methods.

This includes new methods for k-mer distance calculation, structures for k-mer constraints, and SIMD-based batch processing for LCS to improve performance and correctness.
2026-11-03 09:23:37 +11:00

113 lines
3.3 KiB
Go

package obikmer
import (
"math/rand"
"testing"
)
func randomKmer(rng *rand.Rand, k int) uint64 {
return rng.Uint64() & (uint64(1)<<(2*uint(k)) - 1)
}
func TestHammingDistance(t *testing.T) {
rng := rand.New(rand.NewSource(1))
for _, k := range []int{1, 2, 7, 21, 31} {
for n := 0; n < 200; n++ {
a, b := randomKmer(rng, k), randomKmer(rng, k)
sa, sb := DecodeKmer(a, k, nil), DecodeKmer(b, k, nil)
want := 0
for i := range sa {
if sa[i] != sb[i] {
want++
}
}
if got := HammingDistance(a, b, k); got != want {
t.Fatalf("k=%d %s/%s: got %d, want %d", k, sa, sb, got, want)
}
}
}
}
func TestHammingDistanceIgnoresHighBits(t *testing.T) {
if got := HammingDistance(0xFF00, 0x0000, 4); got != 0 {
t.Fatalf("bits above 2k must be ignored, got %d", got)
}
}
func TestKmerLCSKnown(t *testing.T) {
enc := func(s string) uint64 { return EncodeKmer([]byte(s), len(s)) }
cases := []struct {
a, b string
want KmerLCSResult
}{
{"ACGT", "ACGT", KmerLCSResult{4, 4}},
{"AAAA", "CCCC", KmerLCSResult{0, 4}},
{"ACGT", "ACGA", KmerLCSResult{3, 4}},
{"ACGT", "CGTA", KmerLCSResult{3, 5}}, // one shift: 1 indel on each side
}
for _, c := range cases {
if got := KmerLCS(enc(c.a), enc(c.b), 4); got != c.want {
t.Errorf("%s/%s: got %+v, want %+v", c.a, c.b, got, c.want)
}
}
if got := KmerLCS(0, 0, 32); got.LCS != -1 {
t.Errorf("k=32 must be rejected, got %+v", got)
}
}
// The batch implementation (SIMD when built with GOEXPERIMENT=simd) must agree
// with the scalar reference for all k and for batch sizes around the lane count.
func TestKmerLCSBatchMatchesReference(t *testing.T) {
rng := rand.New(rand.NewSource(2))
for _, k := range []int{1, 2, 3, 5, 8, 16, 21, 30, 31} {
for _, n := range []int{0, 1, 3, 7, 8, 9, 16, 17, 100} {
query := randomKmer(rng, k)
partners := make([]uint64, n)
for i := range partners {
partners[i] = randomKmer(rng, k)
}
if n > 2 {
partners[0] = query // identical
}
out := make([]KmerLCSResult, n)
if !KmerLCSBatch(query, partners, k, out) {
t.Fatalf("k=%d n=%d: batch refused", k, n)
}
for i, p := range partners {
if want := KmerLCS(query, p, k); out[i] != want {
t.Fatalf("k=%d n=%d partner %d: got %+v, want %+v", k, n, i, out[i], want)
}
}
}
}
}
func TestKmerLCSBatchRejectsBadInput(t *testing.T) {
if KmerLCSBatch(0, []uint64{1}, 0, make([]KmerLCSResult, 1)) {
t.Fatal("k=0 must be rejected")
}
if KmerLCSBatch(0, []uint64{1, 2}, 4, make([]KmerLCSResult, 1)) {
t.Fatal("short output must be rejected")
}
}
func benchBatch(b *testing.B, k int, f func(uint64, []uint64, int, []KmerLCSResult)) {
rng := rand.New(rand.NewSource(3))
partners := make([]uint64, 1024)
for i := range partners {
partners[i] = randomKmer(rng, k)
}
out := make([]KmerLCSResult, len(partners))
query := randomKmer(rng, k)
b.ResetTimer()
for b.Loop() {
f(query, partners, k, out)
}
b.ReportMetric(float64(b.Elapsed().Nanoseconds())/float64(b.N*len(partners)), "ns/pair")
}
func BenchmarkKmerLCSScalar21(b *testing.B) { benchBatch(b, 21, kmerLCSBatchScalar) }
func BenchmarkKmerLCSBatch21(b *testing.B) { benchBatch(b, 21, kmerLCSBatch) }
func BenchmarkKmerLCSScalar11(b *testing.B) { benchBatch(b, 11, kmerLCSBatchScalar) }
func BenchmarkKmerLCSBatch11(b *testing.B) { benchBatch(b, 11, kmerLCSBatch) }