Protein classification by stochastic modeling and optimal filtering of amino-acid sequences

White, James V.; Stultz, Collin M.; Smith, Temple F.

doi:10.1016/0025-5564(94)90004-3

Cited by 141 publications

(103 citation statements)

References 23 publications

Supporting

Mentioning

102

Contrasting

Order By: Relevance

“…In another paper (White et al, 1992), results on the accuracy of our structural class assignments are reported for study set clusters of homologous sequences (see Table 3 footnote) rather than individual sequences. These results were also obtained with a previous set of DSMs that did not include amphipathic helices or strands.…”

Section: Discussionmentioning

confidence: 99%

Structural analysis based on state‐space modeling

1993

Self Cite

View full text Add to dashboard Cite

A new method has been developed to compute the probability that each amino acid in a protein sequence is in a particular secondary structural element. Each of these probabilities is computed using the entire sequence and a set of predefined structural class models. This set of structural classes is patterned after Jane Richardson's taxonomy for the domains of globular proteins. For each structural class considered, a mathematical model is constructed to represent constraints on the pattern of secondary structural elements characteristic of that class. These are stochastic models having discrete state spaces (referred to as hidden Markov models by researchers in signal processing and automatic speech recognition). Each model is a mathematical generator of amino acid sequences; the sequence under consideration is modeled as having been generated by one model in the set of candidates. The probability that each model generated the given sequence is computed using a filtering algorithm. The protein is then classified as belonging to the structural class having the most probable model. The secondary structure of the sequence is then analyzed using a "smoothing" algorithm that is optimal for that structural class model. For each residue position in the sequence, the smoother computes the probability that the residue is contained within each of the defined secondary structural elements of the model. This method has two important advantages: (1) the probability of each residue being in each of the modeled secondary structural elements is computed using the totality of the amino acid sequence, and (2) these probabilities are consistent with prior knowledge of realizable domain folds as encoded in each model. As an example of the method's utility, we present its application to flavodoxin, a prototypical a//3 protein having a central &sheet, and to thioredoxin, which belongs to a similar structural class but shares no significant sequence similarity.Keywords: flavodoxin; hidden Markov model; secondary structure; state space modeling; tertiary structure classification; thioredoxin The number of known protein sequences greatly exceeds the number of directly determined structures. This is due in part to the relative ease of protein sequence determination by DNA sequencing as compared to the difficulty and expense of structure determination by X-ray crystallography and NMR spectroscopy. Although the number of known structures continues to grow, the number of protein sequences is expected to greatly exceed the number of known structures for the foreseeable future. Because knowledge of the structure of a protein is essential to understanding its function, the elucidation of any aspect of a protein's structure from the sequence information alone is potentially useful.Although there are over 600 structures in the Brookhaven protein structural database (Bernstein et al., 1977;

show abstract

Section: Discussionmentioning

confidence: 99%

Structural analysis based on state‐space modeling

1993

Self Cite

View full text Add to dashboard Cite

show abstract

“…That this structural element must play a critical role with regard to the stability and functionality of the molecule, becomes immediately apparent from the nonfunctional Cysl38-Arg mutant described earlier Xie et al, 1992), in which the Cys112-Cys138 bridge cannot be formed. A prediction of the secondary structure of GM2AP using the type-1 Discrete Space Model of Stultz, White and Smith (Stultz et al, 1993;White et al, 1994) for soluble, monomeric, globular proteins with no known sequence homologues, classifies GM2AP as belonging to the beta superclass ( p = 0.94), typical members of which are, e.g., concanavalin A, macromycin, and rhizopuspepsin. According to this prediction, 81 of the 170 residues of the activator precursor (His24-Ile193) have high probabilities ( p > 0.5) to be included in @-strands, while the rest may form @-turns and irregular regions, with very low probabilities for a-helix formation all over the sequence.…”

Section: Discussionmentioning

confidence: 99%

“…A prediction of secondary structure was run on the PSA server (http://bmerc-www.bu.edu/psa) at the Biomolecular Research Engineering Center, Boston, using the type-I Discrete Space Model (Stultz et al, 1993;White et al, 1994) for monomeric, singledomain, globular, and water-soluble proteins.…”

Section: Cyanylation Of Nascent Sulfhydrylsmentioning

confidence: 99%

Complete localization of disulfide bonds in GM2 activator protein

Schutte¹,

Lemm²,

Glombitza³

et al. 1998

Protein Science

View full text Add to dashboard Cite

Lysosomal degradation of ganglioside GM2 by hexosaminidase A requires the presence of a small, non-enzymatic cofactor, the GMZactivator protein (GM2AP). Lack of functional protein leads to the AB variant of GM2-gangliosidosis, a fatal lysosomal storage disease.Although its possible mode of action and functional domains have been discussed frequently in the past, no structural information about GM2AP is available so far. Here, we determine the complete disulfide bond pattern of the protein. Two of the four disulfide bonds present in the protein were open to classical determination by enzymatic cleavage and mass spectrometry. The direct localization of the remaining two bonds was impeded by the close vicinity of cysteines 136 and 138. We determined the arrangement of these disulfide bonds by MALDI-PSD analysis of disulfide linked peptides and by partial reduction, cyanylation and fragmentation in basic solution, as described recently (Wu F, Watson JT, 1997, Protein Sci 6:391-398).

show abstract

“…Applications of this paradigm range from control systems (Kitagawa and Gersch 1996;Mendel 1995), speech recognition (Becchetti and Ricotti 1999), studies of protein structure (White et al 1994), analysis of genetic networks (Harbison et al 2004), studies of neural representations of biological signals (Brown et al 1998;Barbieri et al 2004;Eden et al 2004;Smith and Brown 2003) and analyses of learning (Wirth et al 2003;Smith et al 2004). A state-space model consists of two equations: a state equation and an observation equation.…”

Section: Introductionmentioning

confidence: 99%

“…When the state and observation models are both linear and Gaussian, the well-known Kalman filter gives the optimal estimate of the unobserved states (Mendel 1995). Extensions of this filter to nonlinear system and/or nonlinear observation models have been widely studied using a range of computational strategies (Fahrmeir and Tutz 2002;Kitagawa and Gersch 1996;White et al 1994;Doucet et al 2001). …”

Section: Introductionmentioning

confidence: 99%

A mixed filter algorithm for cognitive state estimation from simultaneously recorded continuous and binary measures of performance

et al. 2008

View full text Add to dashboard Cite

Continuous (reaction times) and binary (correct/incorrect responses) measures of performance are routinely recorded to track the dynamics of a subject's cognitive state during a learning experiment. Current analyses of experimental data from learning studies do not consider the two performance measures together and do not use the concept of the cognitive state formally to design statistical methods. We develop a mixed filter algorithm to estimate the cognitive state modeled as a linear stochastic dynamical system from simultaneously recorded continuous and binary measures of performance. The mixed filter algorithm has the Kalman filter and the more recently developed recursive filtering algorithm for binary processes as special cases. In the analysis of a simulated learning experiment the mixed filter algorithm provided a more accurate and precise estimate of the cognitive state process than either the Kalman or binary filter alone. In the analysis of an actual learning experiment in which a monkey's performance was tracked by its series of reaction times, and correct and incorrect responses, the mixed filter gave a more complete description of the learning NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author Manuscript process than either the Kalman or binary filter. These results establish the feasibility of estimating cognitive state from simultaneously recorded continuous and binary performance measures and suggest a way to make practical use of concepts from learning theory in the design of statistical methods for the analysis of data from learning experiments.

show abstract

Protein classification by stochastic modeling and optimal filtering of amino-acid sequences

Cited by 141 publications

References 23 publications

Structural analysis based on state‐space modeling

Structural analysis based on state‐space modeling

Complete localization of disulfide bonds in GM2 activator protein

A mixed filter algorithm for cognitive state estimation from simultaneously recorded continuous and binary measures of performance

Contact Info

Product

Resources

About