PS_GG_ALIGN
Profile/Sequence Global Gap Alignment

PS_GG_ALIGN is a FORTRAN90 library which implements some of the string alignment algorithms described in the reference [Chao].

These algorithms carry out the computation in linear space, and compute not just the optimal alignment score, but also the corresponding optimal alignment. The only alignments considered are global, that is, the entirety of both strings are compared. Gaps in the alignment are assigned an affine gap penalty.

It's important to be able to compute alignments using "linear space", that is, just a few vectors whose length N is equal to that of a typical string. A quadratic algorithm would require a two dimensional array of total dimension N*N. Realistic alignment problems can involve strings of N=100,000 elements or more, so a quadratic algorithm would be expensive or impossible to use.

The optimal alignment score is computed without explicitly constructing the corresponding alignment. So a major feature of the algorithms is how to backtrack from the score to retrieve the alignment. It is a matter of some difficulty to recover the matching, particularly if the best score was calculated with a linear space algorithm, which discards a great deal of intermediate information. However, the linear space algorithm implemented here can also compute the optimal matching, based on the idea of a recursive subdivision of the problem.

This set of algorithms does not actually match a pair of sequences, but rather matches a sequence to a "profile". A profile is constructed based on information from many sequences, and can be thought of as a "generalized sequence", or a set of indices, where for each index we specify the likelihood that each possible nucleic acid will occur. These likelihoods can then be used to score the alignments we consider with the new candidate sequence.

This set of routines assumes that an insertion or deletion of length K is penalized using an "affine gap penalty formula" of the form:

Penalty = Gap_Open + K * Gap_Extend

Routines that use quadratic space are included as well, so the algorithms can be compared for storage, speed, and correctness.

Licensing:

The computer code and data files described and made available on this web page are distributed under the GNU LGPL license.

Languages:

PS_GG_ALIGN is available in a FORTRAN90 version.

Related Data and Programs:

PS_LG_ALIGN, a FORTRAN90 library which implements a profile/sequence local alignment using an affine gap penalty.

PS_QG_ALIGN, a FORTRAN90 library which implements a profile/sequence quasiglobal alignment using an affine gap penalty.

SS_GD_ALIGN, a FORTRAN90 library which globally aligns two sequences using a distance matrix.

SS_GG_ALIGN, a FORTRAN90 library which globally aligns two sequences using an affine gap penalty.

SS_LG_ALIGN, a FORTRAN90 library which locally aligns two sequences using an affine gap penalty.

SS_QG_ALIGN, a FORTRAN90 library which quasi-globally aligns two sequences using an affine gap penalty.

Reference:

1. Kun-Mao Chao, Ross Hardison, Webb Miller,
   Recent Developments in Linear-Space Alignment Methods: A Survey,
   Journal of Computational Biology,
   Volume 1, Number 4, 1994, pages 271-291.
2. Eugene Myers, Webb Miller,
   Optimal Alignments in Linear Space,
   CABIOS,
   Volume 4, number 1, 1988, pages 11-17.
3. Albert Nijenhuis, Herbert Wilf,
   Combinatorial Algorithms for Computers and Calculators,
   Second Edition,
   Academic Press, 1978,
   ISBN: 0-12-519260-6,
   LC: QA164.N54.
4. Michael Waterman,
   Introduction to Computational Biology,
   Chapman and Hall, 1995,
   ISBN: 0412993910,
   LC: QH438.4.M33.W38.

Source Code:

• ps_gg_align.f90, the source code;
• ps_gg_align.sh, commands to compile the source code;

Examples and Tests:

• profile.txt, a sample profile scoring data file.
• ps_gg_align_test.f90, a sample program to read a profile scoring data file.
• ps_gg_align_test.sh, commands to compile, link and run the source code.
• ps_gg_align_test.out, output from running the sample program.

List of routines:

• A_INDEX sets up a reverse index for the amino acid codes.
• A_TO_I returns the index of an alphabetic character.
• CH_CAP capitalizes a single character.
• I_SWAP switches two integer values.
• I_TO_A returns the I-th alphabetic character.
• IVEC_REVERSE reverses the elements of an integer vector.
• IVEC2_COMPARE compares pairs of integers stored in two vectors.
• IVEC2_SORT_A ascending sorts a vector of pairs of integers.
• PROFILE_SCORE_PRINT prints profile scoring data.
• PROFILE_SCORE_READ reads profile scoring data from a file.
• PROFILE_SCORE_READ2 returns a small amount of information from a profile.
• PS_GG_BPQ determines a global gap backward alignment path in quadratic space.
• PS_GG_BSL determines a global gap backward alignment score in linear space.
• PS_GG_BSQ determines a global gap backward alignment score in quadratic space.
• PS_GG_FPQ determines a global gap forward alignment path in quadratic space.
• PS_GG_FSL determines a global gap forward alignment score in linear space.
• PS_GG_FSQ determines a global gap forward alignment score in quadratic space.
• PS_GG_MATCH_PRINT prints a global gap alignment.
• PS_GG_RPL determines a global gap recursive alignment path in linear space.
• PS_GG_RPL_POP pops the data describing a subproblem off of the stack.
• PS_GG_RPL_PUSH pushes the data describing a subproblem onto the stack.
• RVEC2_SUM_IMAX returns the index of the maximum sum of two real vectors.
• S_EQI is a case insensitive comparison of two strings for equality.
• S_TO_CVEC converts a string to a character vector.
• S_TO_I reads an integer value from a string.
• SORT_HEAP_EXTERNAL externally sorts a list of items into linear order.
• TIMESTAMP prints the current YMDHMS date as a time stamp.
• WORD_LAST_READ returns the last word from a string.
• WORD_NEXT_READ "reads" words from a string, one at a time.

You can go up one level to the FORTRAN90 source codes.