dna5.hpp

Go to the documentation of this file.

// SPDX-FileCopyrightText: 2006-2024 Knut Reinert & Freie Universität Berlin
// SPDX-FileCopyrightText: 2016-2024 Knut Reinert & MPI für molekulare Genetik
// SPDX-License-Identifier: BSD-3-Clause
 
#pragma once
 
#include <vector>
 
#include <seqan3/alphabet/nucleotide/nucleotide_base.hpp>
#include <seqan3/utility/char_operations/transform.hpp>
 
// ------------------------------------------------------------------
// dna5
// ------------------------------------------------------------------
 
namespace seqan3
{
 
class rna5;
 
class dna5 : public nucleotide_base<dna5, 5>
{
private:
    using base_t = nucleotide_base<dna5, 5>;
 
    friend base_t;
    friend base_t::base_t;
    friend rna5;
 
public:
    constexpr dna5() noexcept = default;                         
    constexpr dna5(dna5 const &) noexcept = default;             
    constexpr dna5(dna5 &&) noexcept = default;                  
    constexpr dna5 & operator=(dna5 const &) noexcept = default; 
    constexpr dna5 & operator=(dna5 &&) noexcept = default;      
    ~dna5() noexcept = default;                                  
 
    using base_t::base_t;
 
    template <std::same_as<rna5> t> // Accept incomplete type
    constexpr dna5(t const & r) noexcept
    {
        assign_rank(r.to_rank());
    }
 
private:
    static constexpr char_type rank_to_char_table[alphabet_size]{'A', 'C', 'G', 'N', 'T'};
 
    static constexpr rank_type rank_complement_table[alphabet_size]{
        4, // T is complement of 'A'_dna5
        2, // G is complement of 'C'_dna5
        1, // C is complement of 'G'_dna5
        3, // N is complement of 'N'_dna5
        0  // A is complement of 'T'_dna5
    };
 
    static constexpr rank_type rank_complement(rank_type const rank)
    {
        return rank_complement_table[rank];
    }
 
    static constexpr char_type rank_to_char(rank_type const rank)
    {
        return rank_to_char_table[rank];
    }
 
    static constexpr rank_type char_to_rank(char_type const chr)
    {
        using index_t = std::make_unsigned_t<char_type>;
        return char_to_rank_table[static_cast<index_t>(chr)];
    }
 
    // clang-format off
    static constexpr std::array<rank_type, 256> char_to_rank_table
    {
        []() constexpr {
            std::array<rank_type, 256> ret{};
 
            ret.fill(3u); // initialize with UNKNOWN ('N')
 
            // reverse mapping for characters and their lowercase
            for (size_t rnk = 0u; rnk < alphabet_size; ++rnk)
            {
                ret[rank_to_char_table[rnk]] = rnk;
                ret[to_lower(rank_to_char_table[rnk])] = rnk;
            }
 
            // set U equal to T
            ret['U'] = ret['T'];
            ret['u'] = ret['t'];
 
            // iupac characters are implicitly "UNKNOWN"
            return ret;
        }()
    };
};
// clang-format on
 
// ------------------------------------------------------------------
// containers
// ------------------------------------------------------------------
 
using dna5_vector = std::vector<dna5>;
 
// ------------------------------------------------------------------
// literals
// ------------------------------------------------------------------
inline namespace literals
{
 
constexpr dna5 operator""_dna5(char const c) noexcept
{
    return dna5{}.assign_char(c);
}
 
SEQAN3_WORKAROUND_LITERAL dna5_vector operator""_dna5(char const * s, std::size_t n)
{
    dna5_vector r;
    r.resize(n);
 
    for (size_t i = 0; i < n; ++i)
        r[i].assign_char(s[i]);
 
    return r;
}
 
} // namespace literals
 
} // namespace seqan3