Parameter a is a harmonic coefficient in integrating dynamic local average connectivity and complex centrality. Table 4 shows the effect of parameter a on the performance of CDLC algorithms, where k is the number of top ranked proteins. CDLC can predict the highest number of true essential proteins with a= 0.1. This observation suggests that the complex component is an excellent property for predicting essential proteins while the LAC component only has a little effect. This finding can be attributed to the fact that essentiality is a product of protein complexes rather
than the individual protein, and protein complexes have a high collection of essential proteins (Hart et al., 2007). Nevertheless, comparing the result of CDLC with a= 0 and that with a= 0.1, we find that a better result can be achieved by the latter value of a. In addition, the protein complexes of some species are less completed than yeast. Therefore, deleting the LAC component may significantly reduce prediction precision. Thus, the LAC component is indispensable although it only has a little effect on the results for yeast data set.
Parameter a is a harmonic coefficient in integrating dynamic local average connectivity and complex centrality. Table 4 shows the effect of parameter a on the performance of CDLC algorithms, where k is the number of top ranked proteins. CDLC can predict the highest number of true essential proteins with a= 0.1. This observation suggests that the complex component is an excellent property for predicting essential proteins while the LAC component only has a little effect. This finding can be attributed to the fact that essentiality is a product of protein complexes rather
than the individual protein, and protein complexes have a high collection of essential proteins (Hart et al., 2007). Nevertheless, comparing the result of CDLC with a= 0 and that with a= 0.1, we find that a better result can be achieved by the latter value of a. In addition, the protein complexes of some species are less completed than yeast. Therefore, deleting the LAC component may significantly reduce prediction precision. Thus, the LAC component is indispensable although it only has a little effect on the results for yeast data set.
การแปล กรุณารอสักครู่..
Parameter a is a harmonic coefficient in integrating dynamic local average connectivity and complex centrality. Table 4 shows the effect of parameter a on the performance of CDLC algorithms, where k is the number of top ranked proteins. CDLC can predict the highest number of true essential proteins with a= 0.1. This observation suggests that the complex component is an excellent property for predicting essential proteins while the LAC component only has a little effect. This finding can be attributed to the fact that essentiality is a product of protein complexes rather
than the individual protein, and protein complexes have a high collection of essential proteins (Hart et al., 2007). Nevertheless, comparing the result of CDLC with a= 0 and that with a= 0.1, we find that a better result can be achieved by the latter value of a. In addition, the protein complexes of some species are less completed than yeast. Therefore, deleting the LAC component may significantly reduce prediction precision. Thus, the LAC component is indispensable although it only has a little effect on the results for yeast data set.
การแปล กรุณารอสักครู่..