A new method for predicting signal sequence cleavage sites is described. The method uses a weight-matrix approach to identify secretory signal sequences and predict cleavage sites with high accuracy. The predictive accuracy is estimated to be around 75-80% for both prokaryotic and eukaryotic proteins, which is a significant improvement over the original method, which was around 65% and 45% accurate for eukaryotic and prokaryotic proteins, respectively. The method involves calculating weight-matrices based on observed amino acid counts in each position of the signal sequences. These matrices are used to scan new sequences and identify the most probable cleavage site. The method was tested on a collection of 161 eukaryotic and 36 prokaryotic signal sequences with known cleavage sites. The results showed that the method correctly identified 78% of eukaryotic and 89% of prokaryotic cleavage sites. The method also allows for efficient discrimination between signal sequences and the N-terminal regions of cytosolic proteins. The method is based on the (-3,-1)-rule, which suggests that the cleavage site is located between positions -3 and -1 of the signal sequence. However, the new method has shown that this rule holds remarkably well, with only a few exceptions. The method also shows that the prokaryotic sequences have stronger constraints in the (-3,-1)-region compared to eukaryotic sequences. The method was tested on a subset of the data to avoid overfitting. The results showed that the method correctly identified the cleavage sites with high accuracy. The method is also able to predict the cleavage site with high accuracy for sequences not included in the data base. The method is simple to implement and can be used on a micro-computer. It provides a clean discrimination between signal sequences and the N-terminal regions of cytosolic proteins. The method is a significant improvement over previous methods and has been shown to be accurate in predicting cleavage sites for both prokaryotic and eukaryotic proteins.A new method for predicting signal sequence cleavage sites is described. The method uses a weight-matrix approach to identify secretory signal sequences and predict cleavage sites with high accuracy. The predictive accuracy is estimated to be around 75-80% for both prokaryotic and eukaryotic proteins, which is a significant improvement over the original method, which was around 65% and 45% accurate for eukaryotic and prokaryotic proteins, respectively. The method involves calculating weight-matrices based on observed amino acid counts in each position of the signal sequences. These matrices are used to scan new sequences and identify the most probable cleavage site. The method was tested on a collection of 161 eukaryotic and 36 prokaryotic signal sequences with known cleavage sites. The results showed that the method correctly identified 78% of eukaryotic and 89% of prokaryotic cleavage sites. The method also allows for efficient discrimination between signal sequences and the N-terminal regions of cytosolic proteins. The method is based on the (-3,-1)-rule, which suggests that the cleavage site is located between positions -3 and -1 of the signal sequence. However, the new method has shown that this rule holds remarkably well, with only a few exceptions. The method also shows that the prokaryotic sequences have stronger constraints in the (-3,-1)-region compared to eukaryotic sequences. The method was tested on a subset of the data to avoid overfitting. The results showed that the method correctly identified the cleavage sites with high accuracy. The method is also able to predict the cleavage site with high accuracy for sequences not included in the data base. The method is simple to implement and can be used on a micro-computer. It provides a clean discrimination between signal sequences and the N-terminal regions of cytosolic proteins. The method is a significant improvement over previous methods and has been shown to be accurate in predicting cleavage sites for both prokaryotic and eukaryotic proteins.