More Like this

1. DTL-DephosSite: Deep Transfer Learning Based Approach to Predict Dephosphorylation Sites

   Chaudhari M, Thapa (June 2021, Frontiers in cell and developmental biology)

   null (Ed.)

   Phosphorylation, which is mediated by protein kinases and opposed by protein phosphatases, is an important post-translational modification that regulates many cellular processes, including cellular metabolism, cell migration, and cell division. Due to its essential role in cellular physiology, a great deal of attention has been devoted to identifying sites of phosphorylation on cellular proteins and understanding how modification of these sites affects their cellular functions. This has led to the development of several computational methods designed to predict sites of phosphorylation based on a protein’s primary amino acid sequence. In contrast, much less attention has been paid to dephosphorylation and its role in regulating the phosphorylation status of proteins inside cells. Indeed, to date, dephosphorylation site prediction tools have been restricted to a few tyrosine phosphatases. To fill this knowledge gap, we have employed a transfer learning strategy to develop a deep learning-based model to predict sites that are likely to be dephosphorylated. Based on independent test results, our model, which we termed DTL-DephosSite, achieved efficiency scores for phosphoserine/phosphothreonine residues of 84%, 84% and 0.68 with respect to sensitivity (SN), specificity (SP) and Matthew’s correlation coefficient (MCC). Similarly, DTL-DephosSite exhibited efficiency scores of 75%, 88% and 0.64 for phosphotyrosine residues with respect to SN, SP, and MCC. 

   more » « less
2. DeepRMethylSite: a deep learning based approach for prediction of arginine methylation sites in proteins

   https://doi.org/10.1039/D0MO00025F  

   Chaudhari, Meenal; Thapa, Niraj; Roy, Kaushik; Newman, Robert H.; Saigo, Hiroto; B. K. C., Dukka (January 2020, Molecular Omics)

   Methylation, which is one of the most prominent post-translational modifications on proteins, regulates many important cellular functions. Though several model-based methylation site predictors have been reported, all existing methods employ machine learning strategies, such as support vector machines and random forest, to predict sites of methylation based on a set of “hand-selected” features. As a consequence, the subsequent models may be biased toward one set of features. Moreover, due to the large number of features, model development can often be computationally expensive. In this paper, we propose an alternative approach based on deep learning to predict arginine methylation sites. Our model, which we termed DeepRMethylSite, is computationally less expensive than traditional feature-based methods while eliminating potential biases that can arise through features selection. Based on independent testing on our dataset, DeepRMethylSite achieved efficiency scores of 68%, 82% and 0.51 with respect to sensitivity (SN), specificity (SP) and Matthew's correlation coefficient (MCC), respectively. Importantly, in side-by-side comparisons with other state-of-the-art methylation site predictors, our method performs on par or better in all scoring metrics tested. 

   more » « less
3. DeepRMethylSite: A Deep Learning Based Approach for prediction of arginine methylation sites in Proteins

   Meenal Chaudhari; Niraj Thapa; Kaushik Roy; Robert H. Newman; Hiroto, Saigod; Dukka B. K. C. (June 2020, Molecular omics)

   Methylation, which is one of the most prominent post-translational modifications on proteins, regulates many important cellular functions. Though several model-based methylation site predictors have been reported, all existing methods employ machine learning strategies, such as support vector machines and random forest, to predict sites of methylation based on a set of “hand-selected” features. As a consequence, the subsequent models may be biased toward one set of features. Moreover, due to the large number of features, model development can often be computationally expensive. In this paper, we propose an alternative approach based on deep learning to predict arginine methylation sites. Our model, which we termed DeepRMethylSite, is computationally less expensive than traditional feature-based methods while eliminating potential biases that can arise through features selection. Based on independent testing on our dataset, DeepRMethylSite achieved efficiency scores of 68%, 82% and 0.51 with respect to sensitivity (SN), specificity (SP) and Matthew's correlation coefficient (MCC), respectively. Importantly, in side-by-side comparisons with other state-of-the-art methylation site predictors, our method performs on par or better in all scoring metrics tested. 

   more » « less
4. A Deep Learning Based Approach for prediction of arginine methylation sites in Proteins

   Chaudhari, Meenal; North Carolina Agricultural and Technical State University, Computational Science; North Carolina Agricultural and Technical State University, Computational Science; North Carolina Agricultural and Technical State University, CS Newman; North Carolina Agricultural and Technical State University, Biology Saigo; Kyushu University, Informatics Dukka; Wichita State University, Electrical Engineering (May 2020, Molecular omics)

   Methylation, which is one of the most prominent post-translational modifications on proteins, regulates many important cellular functions. Though several model-based methylation site predictors have been reported, all existing methods employ machine learning strategies, such as support vector machines and random forest, to predict sites of methylation based on a set of “hand-selected” features. As a consequence, the subsequent models may be biased toward one set of features. Moreover, due to the large number of features, model development can often be computationally expensive. In this paper, we propose an alternative approach based on deep learning to predict arginine methylation sites. Our model, which we termed DeepRMethylSite, is computationally less expensive than traditional feature-based methods while eliminating potential biases that can arise through features selection. Based on independent testing on our dataset, DeepRMethylSite achieved efficiency scores of 68%, 82% and 0.51 with respect to sensitivity (SN), specificity (SP) and Matthew’s correlation coefficient (MCC), respectively. Importantly, in side-by-side comparisons with other state-of-the-art methylation site predictors, our method performs on par or better in all scoring metrics tested. 

   more » « less
5. An Inherent Difference between Serine and Threonine Phosphorylation: Phosphothreonine Strongly Prefers a Highly Ordered, Compact, Cyclic Conformation

   https://doi.org/10.1021/acschembio.3c00068  

   Pandey, Anil K.; Ganguly, Himal K.; Sinha, Sudipta Kumar; Daniels, Kelly E.; Yap, Glenn P.; Patel, Sandeep; Zondlo, Neal J. (September 2023, ACS Chemical Biology)

   Phosphorylation and dephosphorylation of proteins by kinases and phosphatases are central to cellular responses and function. The structural effects of serine and threonine phosphorylation were examined in peptides and in proteins, by circular dichroism, NMR spectroscopy, bioinformatics analysis of the PDB, small-molecule X-ray crystallography, and computational investigations. Phosphorylation of both serine and threonine residues induces substantial conformational restriction in their physiologically more important dianionic forms. Threonine exhibits a particularly strong disorder-to-order transition upon phosphorylation, with dianionic phosphothreonine preferentially adopting a cyclic conformation with restricted φ (φ ~ –60 ̊) stabilized by three noncovalent interactions: a strong intraresidue phosphate-amide hydrogen bond, an n→π* interaction between consecutive carbonyls, and an n→σ* interaction between the phosphate Oγ lone pair and the antibonding orbital of C–Hβ that restricts the χ2 side chain conformation. Proline is unique among the canonical amino acids for its covalent cyclization on the backbone. Phosphothreonine can mimic proline's backbone cyclization via noncovalent interactions. The preferred torsions of dianionic phosphothreonine are φ,ψ = polyproline II helix > α-helix (φ ~ –60 ̊); χ1 = g–; χ2 ~ +115 ̊ (eclipsed C–H/O–P bonds). This structural signature is observed in diverse proteins, including in the activation loops of protein kinases and in protein-protein interactions. In total, these results suggest a structural basis for the differential use and evolution of threonine versus serine phosphorylation sites in proteins, with serine phosphorylation typically inducing smaller, rheostat-like changes, versus threonine phosphorylation promoting larger, step function-like switches, in proteins. 

   more » « less