Prediction of the disulfide-bonding state of cysteine in proteins

Abstract

The bonding states of cysteine play important functional and structural roles in proteins. In particular, disulfide bond formation is one of the most important factors influencing the three-dimensional fold of proteins. Proteins of known structure were used to teach computer-simulated neural networks rules for predicting the disulfide-bonding state of a cysteine given only its flanking amino acid sequence. Resulting networks make accurate predictions on sequences different from those used in training, suggesting that local sequence greatly influences cysteines in disulfide bond formation. The average prediction rate after seven independent network experiments is 81.4% for disulfide-bonded and 80.0% for non-disulfide-bonded scenarios. Predictive accuracy is related to the strength of network output activities. Network weights reveal interesting position-dependent amino acid preferences and provide a physical basis for understanding the correlation between the flanking sequence and a cysteine's disulfide-bonding state. Network predictions may be used to increase or decrease the stability of existing disulfide bonds or to aid the search for potential sites to introduce new disulfide bonds.

Characterization of caseinomacropeptides released from renneted raw and UHT treated milks

“…In other words, it appears that disulfide bridges between cysteinyl residues of both prateins must be strongly infiuenced by the presence of sugars in the environment. Presence of hydraphilic residues araund cysteine favors disulfide bond formation (Muskal et al, 1990) and it can be deduced fram recent 3-dimensional molecular modeling of x-casein molecule (Kumosinski et al, 1991) that 11 and 88 cysteinyl residues are close to 133 threonine residue, the main site of glycosylation.…”

Current advances in disulfide connectivity predictions

“…The overview of the prediction accuracy (Q 2 ) for the prediction of disulfide bonding state with various combinations of features by different methods (Chen et al, 2004;Fiser and Simon, 2000;Fiser et al, 1992;Martelli et al, 2002a;Mucchielli-Giorgi et al, 2002;Muskal et al, 1990;Song et al, 2004). It should be noted that direct comparison of these results may be biased since different datasets and experimental protocols were used.…”

Artificial intelligence and molecular biology

“…Recently, Muskal et al [1990] used a network of the architecture seen in Figure 9 to predict a cysteine's disulfide bonding state, with the presumption that it is the local sequence that influences a cysteine's preference for forming a disulfide bond. The networks were of the feedforward type containing no hidden nodes (perceptrons).…”