Simple dictionary in C++

前端 未结 8 1219

Moving some code from Python to C++.

BASEPAIRS = { \"T\": \"A\", \"A\": \"T\", \"G\": \"C\", \"C\": \"G\" }

Thinking maps might be overkill? W

8条回答
  •  囚心锁ツ
    2021-01-30 16:53

    While using a std::map is fine or using a 256-sized char table would be fine, you could save yourself an enormous amount of space agony by simply using an enum. If you have C++11 features, you can use enum class for strong-typing:

    // First, we define base-pairs. Because regular enums
    // Pollute the global namespace, I'm using "enum class". 
    enum class BasePair {
        A,
        T,
        C,
        G
    };
    
    // Let's cut out the nonsense and make this easy:
    // A is 0, T is 1, C is 2, G is 3.
    // These are indices into our table
    // Now, everything can be so much easier
    BasePair Complimentary[4] = {
        T, // Compliment of A
        A, // Compliment of T
        G, // Compliment of C
        C, // Compliment of G
    };
    

    Usage becomes simple:

    int main (int argc, char* argv[] ) {
        BasePair bp = BasePair::A;
        BasePair complimentbp = Complimentary[(int)bp];
    }
    

    If this is too much for you, you can define some helpers to get human-readable ASCII characters and also to get the base pair compliment so you're not doing (int) casts all the time:

    BasePair Compliment ( BasePair bp ) {
        return Complimentary[(int)bp]; // Move the pain here
    }
    
    // Define a conversion table somewhere in your program
    char BasePairToChar[4] = { 'A', 'T', 'C', 'G' };
    char ToCharacter ( BasePair bp ) {
        return BasePairToChar[ (int)bp ];
    }
    

    It's clean, it's simple, and its efficient.

    Now, suddenly, you don't have a 256 byte table. You're also not storing characters (1 byte each), and thus if you're writing this to a file, you can write 2 bits per Base pair instead of 1 byte (8 bits) per base pair. I had to work with Bioinformatics Files that stored data as 1 character each. The benefit is it was human-readable. The con is that what should have been a 250 MB file ended up taking 1 GB of space. Movement and storage and usage was a nightmare. Of coursse, 250 MB is being generous when accounting for even Worm DNA. No human is going to read through 1 GB worth of base pairs anyhow.

提交回复
热议问题