Accurate prediction of protein structural classes by incorporating PSSS and PSSM into Chou's general PseAAC

Predicting protein structural classes based on complex networks and recurrence analysis

“…The PSSM is converted into matrix by defined evolutionary difference equation and finally a 420 dimensional feature vector is extracted according to the gained matrix. In PSSS-PSePSSM [33], the input of the classifier includes 111 features. Among them, 100 features are selected based on pseudo-position specific scoring matrix (PSePSSM) which contain evolutionary information and sequence order information.…”

Prediction of Protein Structural Classes for Low-Similarity Sequences Based on Consensus Sequence and Segmented PSSM

“…We select the accuracy of each class and overall accuracy as evaluation indexes that are summarized in Table 5 . The compared methods include other competitive PSSM-based methods such as PSSM-S [ 36 ], LCC-PSSM [ 25 ], MBMGAC-PSSM [ 40 ], RPSSM [ 34 ], AADP-PSSM [ 15 ], AAC-PSSM-AC [ 17 ], AATP [ 33 ], PsePSSM [ 41 ], Xia et al [ 42 ], and MEDP [ 35 ], which are recently reported protein structural classes prediction methods based on the evolutionary information represented in the form of PSSM. MBMGAC-PSSM is our other method by fusing three autocorrelation descriptors and PSSM.…”

“…Furthermore, we propose 20 composition moment features for CS, which have been applied for prediction of protein structural class mainly based on amino acid sequence [ 47 ] and predicted protein secondary structure sequence [ 34 , 41 ]. They are formulated as where n i is the total number of the i th amino acid of 20 amino acids in the consensus sequence (CS) and n ij represents the j th position in the CS (the length of L ) of amino acid i .…”

“…However, the last segment may be longer or shorter owing to L not being always divisible by n and the last segment length can be L − (( n − 1) ∗L i ). Then, for each segment, we adopt the pseudo-PSSM (PsePSSM), which has been successfully applied to prediction of protein structural class [ 41 ]. Because the length of the shortest sequence of the three datasets is 10 (for 1189 dataset), hence n can be taken to only 2, 3, 4, and 5.…”

Historical Criteria for Structural Classes of Proteins in Percentages: After 20 Years

“…The number of residues inβ-strands that form anti-parallel (Apr)β-sheets, which are designed toimprove the prediction accuracy of proteins from α + βclass does not do so, instead causes confusion, since there are more anti-parallel (Apr)β-sheets in all βclass than α + β class. In addition to helix, sheet, and coin contents of proteins, recently developed features, especially inclusion of features derived from PSSM matrix, improvedthe prediction accuracy of protein classes enormously (Zhang, et al 2011;Zhang, et al 2014;Zhang, 2015;Ding, et al 2014;Liu, and Jia, 2010;Liu, et. al., 2012).…”