aa20.hpp

Go to the documentation of this file.

// SPDX-FileCopyrightText: 2006-2025 Knut Reinert & Freie Universität Berlin
// SPDX-FileCopyrightText: 2016-2025 Knut Reinert & MPI für molekulare Genetik
// SPDX-License-Identifier: BSD-3-Clause

 
#pragma once
 
#include <vector>
 
#include <seqan3/alphabet/aminoacid/aminoacid_base.hpp>
#include <seqan3/alphabet/aminoacid/concept.hpp>
#include <seqan3/utility/char_operations/transform.hpp>
 
namespace seqan3
{

class aa20 : public aminoacid_base<aa20, 20>
{
private:
    using base_t = aminoacid_base<aa20, 20>;

    friend base_t;
    friend base_t::base_t;
 
public:
    constexpr aa20() noexcept = default;                         
    constexpr aa20(aa20 const &) noexcept = default;             
    constexpr aa20(aa20 &&) noexcept = default;                  
    constexpr aa20 & operator=(aa20 const &) noexcept = default; 
    constexpr aa20 & operator=(aa20 &&) noexcept = default;      
    ~aa20() noexcept = default;                                  
 
    using base_t::base_t;
 
private:
    static constexpr char_type rank_to_char_table[alphabet_size]{'A', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'K', 'L',
                                                                 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'V', 'W', 'Y'};

    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)];
    }

    static constexpr std::array<rank_type, 256> char_to_rank_table{
        []() constexpr
        {
            std::array<rank_type, 256> ret{};
 
            // initialize with 'S' because that appears most frequently
            ret.fill(15u);
 
            // reverse mapping for characters and their lowercase
            for (rank_type rnk = 0u; rnk < alphabet_size; ++rnk)
            {
                ret[static_cast<rank_type>(rank_to_char_table[rnk])] = rnk;
                ret[static_cast<rank_type>(to_lower(rank_to_char_table[rnk]))] = rnk;
            }
 
            ret['B'] = ret['D'];
            ret['b'] = ret['D']; // Convert b (either D/N) to D, since D occurs more frequently.
            ret['J'] = ret['L'];
            ret['j'] = ret['L']; // Convert j (either I/L) to L, since L occurs more frequently.
            ret['O'] = ret['L'];
            ret['o'] = ret['L']; // Convert Pyrrolysine to lysine.
            ret['U'] = ret['C'];
            ret['u'] = ret['C']; // Convert Selenocysteine to cysteine.
            ret['X'] = ret['S'];
            ret['x'] = ret['S']; // Convert unknown amino acids to serine.
            ret['Z'] = ret['E'];
            ret['z'] = ret['E']; // Convert z (either E/Q) to E, since E occurs more frequently.
            ret['*'] = ret['W']; // The most common stop codon is UGA. This is most similar to a Tryptophan.
            return ret;
        }()};
};
 
// ------------------------------------------------------------------
// containers
// ------------------------------------------------------------------

using aa20_vector = std::vector<aa20>;
 
// ------------------------------------------------------------------
// literals
// ------------------------------------------------------------------
inline namespace literals
{

constexpr aa20 operator""_aa20(char const c) noexcept
{
    return aa20{}.assign_char(c);
}

SEQAN3_WORKAROUND_LITERAL aa20_vector operator""_aa20(char const * const s, size_t const n)
{
    aa20_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