The second method entailed an initial iterated PSI-BLAST [ 34] search against the NCBI nonredundant protein database to construct a position-specific scoring matrix for each exon variant.
To construct a position-specific scoring matrix, a PGP-BLAST search was carried out against 3 506 454 extracted peptide sequences, with an E-value threshold of 0.002 and four iterations.
Probabilities for all possible substitutions at each position are then computed to construct a position-specific scoring matrix, where each entry in the matrix represents the probability p ca of observing amino acid a in column c in the alignment.
Position-specific-iterated BLAST (PSI-BLAST) iteratively searches a protein sequence database, using the matches in round i to construct a position-specific score matrix (PSSM) for searching the database in round i + 1.
The nodes organize into collaborative subtrees, then using simple geometric relationships each node constructs a position estimate and finally the estimate is refined using a least-squares method.
Briefly, we constructed a position weight matrix (PWM) for NRF2 (ARE motif) based on the collection of experimentally discovered AREs.
We have constructed a position weight matrix (PWM) based tool named sRNAscanner, using E. coli K-12 MG1655 sRNA-specific transcriptional signals as positive training data, for the identification of intergenic sRNAs.
For each of these subsets, we extracted all sequences of length 20 nt covering the region [−10,+10], and for each of the 16 such subsets we constructed a position weight matrix (PWM) following [36].
They constructed a position-specific scoring matrix (PSSM), and observed that the preferred residues for µ- calpain and m- calpain recognition are Leu, Thr and Val in the P2 position, and Lys, Tyr and Arg in the P1 position [16].
