IDP–CRF: Intrinsically Disordered Protein/Region Identification Based on Conditional Random Fields

Abstract

Accurate prediction of intrinsically disordered proteins/regions is one of the most important tasks in bioinformatics, and some computational predictors have been proposed to solve this problem. How to efficiently incorporate the sequence-order effect is critical for constructing an accurate predictor because disordered region distributions show global sequence patterns. In order to capture these sequence patterns, several sequence labelling models have been applied to this field, such as conditional random fields (CRFs). However, these methods suffer from certain disadvantages. In this study, we proposed a new computational predictor called IDP–CRF, which is trained on an updated benchmark dataset based on the MobiDB database and the DisProt database, and incorporates more comprehensive sequence-based features, including PSSMs (position-specific scoring matrices), kmer, predicted secondary structures, and relative solvent accessibilities. Experimental results on the benchmark dataset and two independent datasets show that IDP–CRF outperforms 25 existing state-of-the-art methods in this field, demonstrating that IDP–CRF is a very useful tool for identifying IDPs/IDRs (intrinsically disordered proteins/regions). We anticipate that IDP–CRF will facilitate the development of protein sequence analysis.

Identification of Intrinsically Disordered Proteins and Regions by Length-Dependent Predictors Based on Conditional Random Fields

“…and Peng and Kurgan 30 bIDP-FSP-L with the parameters ratio = 1:1 and window size = 13, IDP-FSP-S with the parameters ratio = 1:5 and window size = 11, and IDP-FSP-G with the parameters ratio = 1:2 and window size = 9.cThe results of OnD-CRF were obtained from the web server.Table 3Comparison of Different Predictors on Independent Test Dataset SL329PredictoraSnSpACCMCCRankACCMCCIDP-FSPb0.750.890.8210.6512IDP-CRF 25 0.750.880.8170.6423SPOT-disorder 17 0.670.960.8150.6731SPINE-D 24 0.780.850.8150.6334DISOPRED3 63 ––0.7950.6155DISOPRED2 55 0.690.900.7950.5956OnD-CRF 16 , c0.790.800.7930.5877MD 53 0.660.890.7750.5887PONDR-FIT 54 0.610.910.7600.5599IUPred-long 50 0.600.920.7600.5599MFDp 18 0.880.620.7500.511211DISOClust 64 0.810.700.7550.511111NORSnet 57 0.54…”

“…aIn addition to OnD-CRF, the results of 18 other compared predictors were obtained from Hanson et al., 17 Zhang et al., 24 and Liu et al 25 …”

“…In this study, protein sequences and their corresponding label sequences are used to train the CRF model, which is a conditional probability model to annotate unlabeled protein sequences. In particular, CRFs have been adopted and proven to be effective for predicting IDPs/IDRs 25 …”

“…Furthermore, the better performance of the second-category predictors shows that these length-dependent predictors are complementary. According to the comparison results presented in a recent review paper, 1 the sequence labeling methods outperform the classification methods, and the latest proposed IDP-conditional random field (CRF) predictor 25 based on CRFs achieves state-of-the-art performance.…”

iEsGene-ZCPseKNC: Identify Essential Genes Based on Z Curve Pseudo $k$ -Tuple Nucleotide Composition

“…As demonstrated by previous studies, such as protein remote homology detection [24]- [26], protein fold recognition [27]- [30], protein disordered region identification [31], [32], DNA binding protein identification [33]- [35] and enhancer prediction [36], the fusion features or predictors can improve the predictive performance.…”