Amino acid substitution matrices.

By Steven Henikoff, Jorja G. Henikoff. 2000

484 citations recorded by the source corpus.

Subfields: Biochemistry, Clinical Biochemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biochemistry

In Clinical Biochemistry

In Molecular Biology

Citations in the corpus

NetMHCpan, a method for MHC class I binding prediction beyond humans (2008, unknown): “…similarity score of the HLA pseudo-sequences using the relation d ¼ 1 s A;Bð Þffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi s A;Að Þ s B;Bð Þ p , where s(A,B) is the BLO- SUM50 similarity score vector (Henikoff and Henikoff 1992) between the pseudo-sequences A and B, respectively.”
Hidden Markov models for detecting remote protein homologies (1998, methodology): “This relative e n tropy measure has been used previously to characterize substitution matrices Altschul, 1991, and the popular BLO- SUM50 and BLOSUM62 matrices corresponds to saving about 0.5 and 0.7 bits per column.”
STRUM: structure-based prediction of protein stability changes upon single-point mutation (2016, background): “…types at the i th position, q ij is the number of times of A ij appearing at the i th position. c ia in Equation (2) is the pseudo-count designed to off-set the deficiency of statistics by where B ( t, a ) is the mutation probability from amino acid t to a in BLOSUM62 (Henikoff and Henikoff, 1992).”
Genome alignment, evolution of prokaryotic genome organization, and prediction of gene function using genomic context. (2001, methodology): “This is generally analogous to the use of a simple identity matrix versus a residue substitution matrix such as PAM or BLOSUM in protein sequence alignment (Henikoff and Henikoff 2000), but, unlike the sequence case, the genome alignments produced using these two approaches differ not only in extent”