Struct AntiLexHasher 

pub struct AntiLexHasher<const CANONICAL: bool> { /* private fields */ }

A hash function that compares strings reverse-lexicographically, with the last (most significant) character inverted.

Only supports 2-bit DNA sequences (packed_seq::AsciiSeq and packed_seq::PackedSeq).

The canonical version (with CANONICAL=true) returns the minimum of the forward and reverse-complement hashes. TODO: Test minimum vs maximum.

Implementations

impl<const CANONICAL: bool> AntiLexHasher<CANONICAL>

pub const fn new(k: usize) -> Self

Create a new AntiLexHasher for kmers of length k.

Trait Implementations

impl KmerHasher for AntiLexHasher<false>

const CANONICAL: bool = false

True when the hash function is invariant under reverse-complement.

fn new(k: usize) -> Self

fn k(&self) -> usize

The value of k for this hasher.

fn in_out_mapper_scalar<'s>( &self, seq: impl Seq<'s>, ) -> impl FnMut((u8, u8)) -> u32

A scalar mapper function that should be called with each (in, out) base.

fn in_out_mapper_simd<'s>( &self, seq: impl Seq<'s>, ) -> impl FnMut((u32x8, u32x8)) -> u32x8

A SIMD mapper function that should be called with a (in, out) base per lane.

fn mapper<'s>(&self, seq: impl Seq<'s>) -> impl FnMut(u8) -> u32

Hash a sequence one character at a time. Ignores k.

fn is_canonical(&self) -> bool

Helper function returning Self::CANONICAL.

fn delay(&self) -> Delay

The delay of the ‘out’ character passed to the in_out_mapper functions. Defaults to k-1.

fn in_out_mapper_ambiguous_scalar<'s>( &self, nseq: PackedNSeq<'s>, ) -> impl FnMut((u8, u8)) -> u32

fn in_out_mapper_ambiguous_simd<'s>( &self, nseq: PackedNSeq<'s>, context: usize, ) -> impl FnMut((u32x8, u32x8)) -> u32x8

fn hash_kmers_scalar<'s>( &self, seq: impl Seq<'s>, ) -> impl ExactSizeIterator<Item = u32>

A scalar iterator over all k-mer hashes in seq.

fn hash_kmers_simd<'s>( &self, seq: impl Seq<'s>, context: usize, ) -> PaddedIt<impl ChunkIt<u32x8>>

A SIMD-parallel iterator over all k-mer hashes in seq.

fn hash_valid_kmers_scalar<'s>( &self, nseq: PackedNSeq<'s>, ) -> impl ExactSizeIterator<Item = u32>

An iterator over all k-mer hashes in seq. Ambiguous kmers get hash u32::MAX.

fn hash_valid_kmers_simd<'s, 't>( &'t self, nseq: PackedNSeq<'s>, context: usize, ) -> PaddedIt<impl ChunkIt<u32x8> + use<'s, 't, Self>>

A SIMD-parallel iterator over all k-mer hashes in seq. Ambiguous kmers get hash u32::MAX.

fn hash_seq<'s>(&self, seq: impl Seq<'s>) -> u32

Hash the given sequence. Ignores k.

fn hash_prefixes<'s>( &self, seq: impl Seq<'s>, ) -> impl ExactSizeIterator<Item = u32>

Hash all non-empty prefixes of the given sequence. Ignores k.

impl KmerHasher for AntiLexHasher<true>

const CANONICAL: bool = true

True when the hash function is invariant under reverse-complement.

fn new(k: usize) -> Self

fn k(&self) -> usize

The value of k for this hasher.

fn delay(&self) -> Delay

The delay of the ‘out’ character passed to the in_out_mapper functions. Defaults to k-1.

fn mapper<'s>(&self, seq: impl Seq<'s>) -> impl FnMut(u8) -> u32

Hash a sequence one character at a time. Ignores k.

fn in_out_mapper_scalar<'s>( &self, seq: impl Seq<'s>, ) -> impl FnMut((u8, u8)) -> u32

A scalar mapper function that should be called with each (in, out) base.

fn in_out_mapper_simd<'s>( &self, seq: impl Seq<'s>, ) -> impl FnMut((u32x8, u32x8)) -> u32x8

A SIMD mapper function that should be called with a (in, out) base per lane.

fn is_canonical(&self) -> bool

Helper function returning Self::CANONICAL.

fn in_out_mapper_ambiguous_scalar<'s>( &self, nseq: PackedNSeq<'s>, ) -> impl FnMut((u8, u8)) -> u32

fn in_out_mapper_ambiguous_simd<'s>( &self, nseq: PackedNSeq<'s>, context: usize, ) -> impl FnMut((u32x8, u32x8)) -> u32x8

fn hash_kmers_scalar<'s>( &self, seq: impl Seq<'s>, ) -> impl ExactSizeIterator<Item = u32>

A scalar iterator over all k-mer hashes in seq.

fn hash_kmers_simd<'s>( &self, seq: impl Seq<'s>, context: usize, ) -> PaddedIt<impl ChunkIt<u32x8>>

A SIMD-parallel iterator over all k-mer hashes in seq.

fn hash_valid_kmers_scalar<'s>( &self, nseq: PackedNSeq<'s>, ) -> impl ExactSizeIterator<Item = u32>

An iterator over all k-mer hashes in seq. Ambiguous kmers get hash u32::MAX.

fn hash_valid_kmers_simd<'s, 't>( &'t self, nseq: PackedNSeq<'s>, context: usize, ) -> PaddedIt<impl ChunkIt<u32x8> + use<'s, 't, Self>>

A SIMD-parallel iterator over all k-mer hashes in seq. Ambiguous kmers get hash u32::MAX.

fn hash_seq<'s>(&self, seq: impl Seq<'s>) -> u32

Hash the given sequence. Ignores k.

fn hash_prefixes<'s>( &self, seq: impl Seq<'s>, ) -> impl ExactSizeIterator<Item = u32>

Hash all non-empty prefixes of the given sequence. Ignores k.