EMBOSS provides a basic maths library that includes functions for datatype conversion, random number generation, rounding and other basics.

AJAX library files for handling maths are listed in the table (Table 6.36, “AJAX Library Files for Handling Maths”). Library file documentation, including a complete description of datatypes and functions, is available at:

http://emboss.open-bio.org/rel/dev/libs/ |

Library File Documentation | Description |
---|---|

ajmath | General maths functions |

ajvector | Functions for handling vectors |

** ajmath.h/c. **General maths functions for rounding numbers, coordinate conversion, random number generation etc.

** ajvector.h/c. **Data structures and functions for handling vectors in 3D space. The library is not described here and you should see the library documentation online for further information.

Functions are provided for converting one datatype to another. These all have the prefix `ajCvt`

. To interconvert polar and cartesian coordinates, to convert radians to degrees or return a probability given a Gaussian distribution use:

float ajCvtDegToRad (float degrees); void ajCvtPolToRec (float radius, float angle, float* Px, float* Py); float ajCvtRadToDeg (float radians); void ajCvtRecToPol (float x, float y, float* Pradius, float* Pangle); double ajCvtGaussToProb (float mean, float sd, float score);

Sequence positions are often specified by the user on the command line. EMBOSS supports negative values for sequence positions which are interpreted as an offset from the *end* of the sequence. Two functions are provided to convert an EMBOSS sequence position to a true position which, for instance, could be used to index into the sequence itself. The functions return an integer between 0 (the start of the sequence) and the length of the sequence minus 1:

ajint ajCvtSposToPos (ajint len, ajint ipos); ajint ajCvtSposToPosStart (ajint len, ajint imin, ajint ipos);

In both cases, `len`

is the length of the sequence and `ipos`

is the position. A value of `0`

means the start of the sequence whereas negative numbers indicate an offset from the end of the sequence. A minimum relative position (`imin`

) can be specified.

There are three functions for random number generation. These are based on `dprand`

and `sdprand`

and used with the permission of N.M. Maclaren (Copyright, The University of Cambridge). `ajRandomSeed`

initialises random number generation and must be called before `ajRandomNumber`

(returns a random integer between 0-32767) and `ajRandomDouble`

(returns a random double precision value between 0.0 and 1.0):

void ajRandomSeed (void); double ajRandomNumber (void); double ajRandomDouble (void);

Two functions are for rounding off numbers, either a `float`

to an `int`

or a `float`

to a `float`

with a specified number of bits left for cumulative addition:

ajint ajRound (ajint i, ajint round); float ajRoundFloat (float f, ajint nbits);

Various miscellaneous functions are provided, for instance, to calculate a geometric mean, the modulo of two numbers or a 32 or 64 bit SwissProt-style checksum for a protein sequence:

float ajMathGmean (const float* Parr, ajint n); ajint ajMathModulo (ajint i1, ajint i2); ajuint ajMathCrc32 (const AjPStr str); unsigned long long ajMathCrc64 (const AjPStr str);