A modified definition of Sov, a segment‐based measure for protein secondary structure prediction assessment

By Adam T. Zemla, Č. Venclovas, Krzysztof Fidelis, Burkhard Rost. 1999

We present a measure for the evaluation of secondary structure prediction methods that is based on secondary structure segments rather than individual residues. The algorithm is an extension of the segment overlap measure Sov, originally defined by Rost et al. (J Mol Biol 1994;235:13–26). The new definition of Sov corrects the normalization procedure and improves Sov's ability to discriminate between similar and dissimilar segment distributions. The method has been comprehensively tested during the second Critical Assessment of Techniques for Protein Structure Prediction (CASP2). Here, we describe the underlying concepts, modifications to the original definition, and their significance. Proteins 1999;34:220–223. Published 1999 Wiley‐Liss, Inc.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction (2014, background): “The Sov scoring function takes segments of continuous structure types into account in an attempt to reward the appropriate placement of predicted structural elements without harshly penalizing individual residuemismatches, especially at the boundaries of segments [30].”
ASTRO-FOLD: a combinatorial and global optimization framework for Ab initio prediction of three-dimensional structures of proteins from the amino acid sequence. (2003, background): “In addition, the SOV analysis is used to compare the ASTRO-FOLD results to those obtained using the PSIPRED method for secondary structure prediction (McGuffin et al., 2000).”
Cascaded multiple classifiers for secondary structure prediction (2000, unknown): “In the following formula, S~i ! denotes a pair of overlapping segments ~Sobs, Spred! in conformation i [ ~H, E, C!, S9~i ! denotes the set of all segment Sobs for which there is no overlapping segment Spred in state i ~for further details see Zemla et al., 1999 !.”
Sequence-based determinants and prediction of fuzzy interactions in protein complexes. (2020, methodology): “Quality assessment e segmental overlap measure Prediction accuracy was assessed using the segmental overlap measure (SOV) [60]: where S obs and S pred are the observed and predicted segments, minov(S obs ;S pred ) is the overlap between S obs and S pred , maxov(S obs ;S pred ) is the total extent…”
A novel method for protein secondary structure prediction using dual‐layer SVM and profiles (2004, methodology): “The three-state overall perresidue accuracy (Q3), the Matthew’s correlation coefficients (CH, CE, CC), and the SOV were used to evaluate the accuracy.(10,22,23) The per-residue accuracies for each type of secondary structure (QH, QE, QC, QH , QE , QC ) were also calculated.”
Secondary structure prediction with support vector machines (2003, methodology): “Table shows the three-state accuracy Q3 (Rost and Sander, 1993) and segment overlap (Sov) scores (Zemla et al., 1999) for a test set of 75”
Predicting transmembrane beta-barrels in proteomes. (2004, methodology): “This problem is addressed by scores that measure the per-segment performance (57,58).”
The effect of long‐range interactions on the secondary structure formation of proteins (2005, unknown): “a The modified Sov segment-based measure (Zemla et al. 1999).”
Critical assessment of high-throughput standalone methods for secondary structure prediction (2011, methodology): “We also compute the segment overlap values that quantify the amount of overlap between the native and the predicted helix (SOVH), strand (SOVE) and coil (SOVC) segments, including the overall segment overlap that consider the three SSs (SOV3) [42].”
Learning sparse models for a dynamic Bayesian network classifier of protein secondary structure (2011, methodology): “The results of the seven-fold cross-validation are summarized in Table 1, including the amino acid level accuracy (called Q 3 [24]), the segment overlap score (SOV) [25], and Matthew ’ s correlation coefficients (MCC) [26].”
EVA: evaluation of protein structure prediction servers (2003, methodology): “EVA-sec measures accuracy by: (i) per-residue accuracy (18) (Q3)—percentage of residues correctly predicted in one of the three states (helix, strand or other); (ii) per-segment accuracy (18, 19 ) (SOV)—average overlap between segments (methods that get most of the segment cores right are generally ”