PARTIAL ALIGNMENT

The partial alignment algorithm is based on the idea of align between two sequences the most preserved regions. This regions use to have the maximum alignment score punctuation. Once the first conserved region is aligned, other regions also conserved but with less score are aligned, and so on. At the end, the alignment is completed by align the rest of the elements.

This kind of alignment is quite usefull in the comparison of functional related sequences.


GLOBAL ALIGNMENT

The global alignment algorithms, like the one that SEQAL applies, tryes to align two sequences from the total score of its alignment. If you want to know more about it see METHODS.

This kind of alignment is usefull in the comparison of evolutionary related sequences.


WEIGH MATRIX

This matrix shows the probability that an specyfic aminoacid change appears during evolution.

PAM MATRIX

Derived from Dayhoff's paper "A Model In Evolutionary Change Proteins". Those matrix represents the logarithm of the likelihood that a change between two aminoacids occurs. Due to this, the negative values represents unlikely changes while positive values represents possible changes. The value expresed with the PAM matrix refers to the quantity of mutations that a concrete position may have had. As an example, PAM250 involves that each residue may have change 2.5 times.


BLOSUM MATRIX

The "BLOcks SUbstitution Matrix" were proposed in 1992 by Steven Henikoff and Jorja G. Henikoff. They were created due to an study of conserved blocks.
Procedure to construct a Blosum matrix:

  1. It starts with conserved sequence's blocs:
    • Align bowth sequences without the option of adding gaps.
  2. Stablish the number of aligned aminoacids' couples (fij).
  3. The observed frequency for each aminoacid couple (qij) is the quotient between the fij value and the total number of aminoacid couples (that includes the couples i=j, were no substitution appears).
  4. The predicted frequency of a couple of aminoacids is the product of the frequencies of each aminoacid in the data conjoint.
  5. All the sequences of a block are agruped according to their threshold of similarity. As an example, if we use an 80% of similarity we are constructing a BLOSUM80 matrix.


GAP OPENING

Penality that we use in order to align an aminoacid with a initial gap.


GAP EXTENSION

Penality that we use as we extend a gap. Generally it is considered less valuable than gap opening, because we consider that, evolutively, the disappearing or appearing of a group of aminoacids is more likely than the same process in a lot of single ones.