Prediction of nuclear proteins using SVM and HMM models

Background  

The nucleus, a highly organized organelle, plays important role in cellular homeostasis. The nuclear proteins are crucial for chromosomal maintenance/segregation, gene expression, RNA processing/export, and many other processes. Several methods have been developed for predicting the nuclear proteins in the past. The aim of the present study is to develop a new method for predicting nuclear proteins with higher accuracy.     

High-yield expression and purification of a monotopic membrane glycosyltransferase

Membrane proteins are essential to many cellular processes. However, the systematic study of membrane protein structure has been hindered by the difficulty in obtaining large quantities of these proteins. Protein overexpression using Escherichia coli is commonly used to produce large quantities of protein, but usually yields very little membrane protein. Furthermore, optimization of the expressing conditions, as well as the choice of detergent and other buffer components, is thought to be crucial for increasing the yield of stable and homogeneous protein. Herein we report high-yield expression and purification of a membrane-associated monotopic protein, the glycosyltransferase monoglucosyldiacylglycerol synthase (alMGS), in E. coli. Systematic optimization of protein expression was achieved through controlling a few basic expression parameters, including temperature and growth media, and the purifications were monitored using a fast and efficient size-exclusion chromatography (SEC) screening method. The latter method was shown to be a powerful tool for fast screening and for finding the optimal protein-stabilizing conditions. For alMGS it was found that the concentration of detergent was just as important as the type of detergent, and a low concentration of n-dodecyl-β-d-maltoside (DDM) (1× critical micelle concentration) was the best for keeping the protein stable and homogeneous. By using these simply methods to optimize the conditions for alMGS expression and purification, the final expression level increase by two orders of magnitude, reaching 170 mg of pure protein per litre culture.     

PRINTR: Prediction of RNA binding sites in proteins using SVM and profiles

Protein–RNA interactions play a key role in a number of biological processes such as protein synthesis, mRNA processing, assembly and function of ribosomes and eukaryotic spliceosomes. A reliable identification of RNA-binding sites in RNA-binding proteins is important for functional annotation and site-directed mutagenesis. We developed a novel method for the prediction of protein residues that interact with RNA using support vector machine (SVM) and position-specific scoring matrices (PSSMs). Two cases have been considered in the prediction of protein residues at RNA-binding surfaces. One is given the sequence information of a protein chain that is known to interact with RNA; the other is given the structural information. Thus, five different inputs have been tested. Coupled with PSI-BLAST profiles and predicted secondary structure, the present approach yields a Matthews correlation coefficient (MCC) of 0.432 by a 7-fold cross-validation, which is the best among all previous reported RNA-binding sites prediction methods. When given the structural information, we have obtained the MCC value of 0.457, with PSSMs, observed secondary structure and solvent accessibility information assigned by DSSP as input. A web server implementing the prediction method is available at the following URL: .     

Structures of the glycosylphosphatidylinositol membrane anchors from Aspergillus fumigatus membrane proteins

Glycosylphosphatidylinositol (GPI)-anchored proteins have been identified in all eukaryotes. In fungi, structural and biosynthetic studies of GPIs have been restricted to the yeast Saccharomyces cerevisiae. In this article, four GPI-anchored proteins were purified from a membrane preparation of the human filamentous fungal pathogen Aspergillus fumigatus. Using new methodology applied to western blot protein bands, the GPI structures were characterized by ES-MS, fluorescence labeling, HPLC, and specific enzymatic digestions. The phosphatidylinositol moiety of the A. fumigatus GPI membrane anchors was shown to be an inositol-phosphoceramide containing mainly phytosphingosine and monohydroxylated C24:0 fatty acid. In constrast to yeast, only ceramide was found in the GPI anchor structures of A. fumigatus, even for Gel1p, a homolog of Gas1p in S. cerevisiae that contains diacylglycerol. The A. fumigatus GPI glycan moiety is mainly a linear pentomannose structure linked to a glucosamine residue: Manalpha1-3Manalpha1-2Manalpha1-2Manalpha1-6Manalpha1-4GlcN.     

Prediction of amphipathic helix—membrane interactions with Rosetta

Amphipathic helices have hydrophobic and hydrophilic/charged residues situated on opposite faces of the helix. They can anchor peripheral membrane proteins to the membrane, be attached to integral membrane proteins, or exist as independent peptides. Despite the widespread presence of membrane-interacting amphipathic helices, there is no computational tool within Rosetta to model their interactions with membranes. In order to address this need, we developed the AmphiScan protocol with PyRosetta, which runs a grid search to find the most favorable position of an amphipathic helix with respect to the membrane. The performance of the algorithm was tested in benchmarks with the RosettaMembrane, ref2015_memb, and franklin2019 score functions on six engineered and 44 naturally-occurring amphipathic helices using membrane coordinates from the OPM and PDBTM databases, OREMPRO server, and MD simulations for comparison. The AmphiScan protocol predicted the coordinates of amphipathic helices within less than 3Å of the reference structures and identified membrane-embedded residues with a Matthews Correlation Constant (MCC) of up to 0.57. Overall, AmphiScan stands as fast, accurate, and highly-customizable protocol that can be pipelined with other Rosetta and Python applications.     

Mechanism of membrane curvature sensing by amphipathic helix containing proteins

There are several examples of membrane-associated protein domains that target curved membranes. This behavior is believed to have functional significance in a number of essential pathways, such as clathrin-mediated endocytosis, which involve dramatic membrane remodeling and require the recruitment of various cofactors at different stages of the process. This work is motivated in part by recent experiments that demonstrated that the amphipathic N-terminal helix of endophilin (H0) targets curved membranes by binding to hydrophobic lipid bilayer packing defects which increase in number with increasing membrane curvature. Here we use state-of-the-art atomistic simulation to explore the packing defect structure of curved membranes, and the effect of this structure on the folding of H0. We find that not only are packing defects increased in number with increasing membrane curvature, but also that their size distribution depends nontrivially on the curvature, falling off exponentially with a decay constant that depends on the curvature, and crucially that even on highly curved membranes defects large enough to accommodate the hydrophobic face of H0 are never observed. We furthermore find that a percolation model for the defects explains the defect size distribution, which implies that larger defects are formed by coalescence of noninteracting smaller defects. We also use the recently developed metadynamics algorithm to study in detail the effect of such defects on H0 folding. It is found that the comparatively larger defects found on a convex membrane promote H0 folding by several kcal/mol, while the smaller defects found on flat and concave membrane surfaces inhibit folding by kinetically trapping the peptide. Together, these observations suggest H0 folding is a cooperative process in which the folding peptide changes the defect structure relative to an unperturbed membrane.     

Prediction of N-myristoylation modification of proteins by SVM

Attachment of a myristoyl group to NH(2)-terminus of a nascent protein among protein post-translational modification (PTM) is called myristoylation. The myristate moiety of proteins plays an important role for their biological functions, such as regulation of membrane binding (HIV-1 Gag) and enzyme activity (AMPK). Several predictors based on protein sequences alone are hitherto proposed. However, they produce a great number of false positive and false negative predictions; or they cannot be used for general purpose (i.e., taxon-specific); or threshold values of the decision rule of predictors need to be selected with cautiousness. Here, we present novel and taxon-free predictors based on protein primary structure. To identify myristoylated proteins accurately, we employ a widely used machinelearning algorithm, support vector machine (SVM). A series of SVM predictors are developed in the present study where various scales representing physicochemical and biological properties of amino acids (from the AAindex database) are used for numerical transformation of protein sequences. Of the predictors, the top ten achieve accuracies of >98% (the average value is 98.34%), and also the area under the ROC curve (AUC) values of >0.98. Compared with those of previous studies, the prediction accuracies are improved by about 3 to 4%.     

Prediction of cystine connectivity using SVM

One of the major contributors to protein structures is the formation of disulphide bonds between selected pairs of cysteines at oxidized state. Prediction of such disulphide bridges from sequence is challenging given that the possible combination of cysteine pairs as the number of cysteines increases in a protein. Here, we describe a SVM (support vector machine) model for the prediction of cystine connectivity in a protein sequence with and without a priori knowledge on their bonding state. We make use of a new encoding scheme based on physico-chemical properties and statistical features (probability of occurrence of each amino acid residue in different secondary structure states along with PSI-blast profiles). We evaluate our method in SPX (an extended dataset of SP39 (swiss-prot 39) and SP41 (swiss-prot 41) with known disulphide information from PDB) dataset and compare our results with the recursive neural network model described for the same dataset.     

Prediction of membrane proteins in Mycobacterium tuberculosis using a support vector machine algorithm.

We report our finding of linear clustering of signal sequences at the N-terminus of M.tb membrane proteins, directing membrane localization. Although it is widely accepted that membrane proteins have signal peptides at the N-terminus, statistical ensemble analysis of Support Vector Machine prediction results indicate that M.tb membrane proteins have embedded N-terminal sequence patterns beyond the signal peptides previously identified in E. coli. The additional patterns at the N-terminus of M.tb membrane proteins may have correlations to their unique enzymatic functions and unusual characteristics such as membrane interaction in pathogenes.     

Stabilization of membranes upon interaction of amphipathic polymers with membrane proteins

Amphipathic polymers derived from polysaccharides, namely hydrophobically modified pullulans, were previously suggested to be useful as polymeric substitutes of ordinary surfactants for efficient and structure-conserving solubilization of membrane proteins, and one such polymer, 18C(10), was optimized for solubilization of proteins derived from bacterial outer membranes (Duval-Terrie et al. 2003). We asked whether a similar ability to solubilize proteins could also be demonstrated in eukaryotic membranes, namely sarcoplasmic reticulum (SR) fragments, the major protein of which is SERCA1a, an integral membrane protein with Ca(2+)-dependent ATPase and Ca(2+)-pumping activity. We found that 18C(10)-mediated solubilization of these SR membranes did not occur. Simultaneously, however, we found that low amounts of this hydrophobically modified pullulan were very efficient at preventing long-term aggregation of these SR membranes. This presumably occurred because the negatively charged polymer coated the membranous vesicles with a hydrophilic corona (a property shared by many other amphipathic polymers), and thus minimized their flocculation. Reminiscent of the old Arabic gum, which stabilizes Indian ink by coating charcoal particles, the newly designed amphipathic polymers might therefore unintentionally prove useful also for stabilization of membrane suspensions.     

High-efficacy subcellular micropatterning of proteins using fibrinogen anchors

Protein micropatterning allows proteins to be precisely deposited onto a substrate of choice and is now routinely used in cell biology and in vitro reconstitution. However, drawbacks of current technology are that micropatterning efficiency can be variable between proteins and that proteins may lose activity on the micropatterns. Here, we describe a general method to enable micropatterning of virtually any protein at high specificity and homogeneity while maintaining its activity. Our method is based on an anchor that micropatterns well, fibrinogen, which we functionalized to bind to common purification tags. This enhances micropatterning on various substrates, facilitates multiplexed micropatterning, and dramatically improves the on-pattern activity of fragile proteins like molecular motors. Furthermore, it enhances the micropatterning of hard-to-micropattern cells. Last, this method enables subcellular micropatterning, whereby complex micropatterns simultaneously control cell shape and the distribution of transmembrane receptors within that cell. Altogether, these results open new avenues for cell biology.     

Anchoring proteins to Escherichia coli cell membranes using hydrophobic anchors derived from a Bacillus subtilis integral membrane protein

Anchored periplasmic expression (APEx) technology aims to express and localize proteins or peptides in the Escherichia coli periplasm. Some reports have suggested that transmembrane segments of integral membrane proteins can be used as membrane anchors in the APEx system. In this study, a series of hydrophobic anchors derived from the first putative transmembrane helix of a Bacillus subtilis integral membrane protein, MrpF, and its truncated forms were investigated for anchored periplasmic expression of alkaline phosphatase (PhoA) in E. coli. Anchoring efficiency of hydrophobic anchors was evaluated by monitoring the expression and activity of anchored PhoA. The length of hydrophobic anchors was found to be critical for anchoring proteins to cell membranes. This study may open new avenues for applying transmembrane segments derived from native membrane proteins as membrane anchors in the APEx system.     

Studies on the amphipathic nature of the membrane proteins in Semliki Forest virus

The membrane glycoproteins E₁ and E₂ of Semliki Forest virus form spikes protruding from the external surface of the virion. They have been cleaved off by thermolysin or subtilisin leaving peptide segments in the membrane of the spikeless virus particles with a molecular weight of about 5000 enriched in hydrophobic amino acids. These peptides are soluble in chloroform/methanol and are solubilized into mixed micelles with Triton X100, with sodium dodecyl sulphate and with sodium deoxycholate. Peptide mapping studies show that each membrane glycoprotein has its own lipophilic peptide segment which presumably serves to anchor these proteins to the lipid membrane. The hydrophobic segments of the glycoproteins appear to be shielded from proteolysis not only by the lipids in the intact membrane but also by Triton X100 in the detergent-protein complexes obtained when this detergent is used to remove the lipid and solubilize the proteins.     

The role of lipid anchors for small G proteins in membrane trafficking

Small GTP-binding proteins of the Ras superfamily play diverse roles in intracellular trafficking. In order to perform these functions, the proteins must associate with specific donor vesicles and be recycled after fusion of these vesicles with their acceptor membrane target. Recent results have identified a number of lipid modifications of these proteins, occurring at the N- or C-termini, that contribute to their membrane binding. Recycling appears, in some cases, to be mediated by soluble proteins that bind the lipid-modified tails, removing them from the membrane and allowing their reutilization via the cytosol.     

Prediction of membrane proteins using split amino acid and ensemble classification

Knowledge of the types of membrane protein provides useful clues in deducing the functions of uncharacterized membrane proteins. An automatic method for efficiently identifying uncharacterized proteins is thus highly desirable. In this work, we have developed a novel method for predicting membrane protein types by exploiting the discrimination capability of the difference in amino acid composition at the N and C terminus through split amino acid composition (SAAC). We also show that the ensemble classification can better exploit this discriminating capability of SAAC. In this study, membrane protein types are classified using three feature extraction and several classification strategies. An ensemble classifier Mem-EnsSAAC is then developed using the best feature extraction strategy. Pseudo amino acid (PseAA) composition, discrete wavelet analysis (DWT), SAAC, and a hybrid model are employed for feature extraction. The nearest neighbor, probabilistic neural network, support vector machine, random forest, and Adaboost are used as individual classifiers. The predicted results of the individual learners are combined using genetic algorithm to form an ensemble classifier, Mem-EnsSAAC yielding an accuracy of 92.4 and 92.2% for the Jackknife and independent dataset test, respectively. Performance measures such as MCC, sensitivity, specificity, F-measure, and Q-statistics show that SAAC-based prediction yields significantly higher performance compared to PseAA- and DWT-based systems, and is also the best reported so far. The proposed Mem-EnsSAAC is able to predict the membrane protein types with high accuracy and consequently, can be very helpful in drug discovery. It can be accessed at http://111.68.99.218/membrane.     

Prediction of protein B-factors using multi-class bounded SVM

In this paper, we propose the adoption of the bounded support vector machine (BSVM) to predict the B-factors of residues based on a number of distinctive properties of residues. Due to the ability of multi-class classification of the BSVM, we can elaborately distinguish our targets and obtain relatively higher accuracy.     

Functional assignment to JEV proteins using SVM

Identification of different protein functions facilitates a mechanistic understanding of Japanese encephalitis virus (JEV) infection and opens novel means for drug development. Support vector machines (SVM), useful for predicting the functional class of distantly related proteins, is employed to ascribe a possible functional class to Japanese encephalitis virus protein. Our study from SVMProt and available JE virus sequences suggests that structural and nonstructural proteins of JEV genome possibly belong to diverse protein functions, are expected to occur in the life cycle of JE virus. Protein functions common to both structural and non-structural proteins are iron-binding, metal-binding, lipid-binding, copper-binding, transmembrane, outer membrane, channels/Pores - Pore-forming toxins (proteins and peptides) group of proteins. Non-structural proteins perform functions like actin binding, zinc-binding, calcium-binding, hydrolases, Carbon-Oxygen Lyases, P-type ATPase, proteins belonging to major facilitator family (MFS), secreting main terminal branch (MTB) family, phosphotransfer-driven group translocators and ATP-binding cassette (ABC) family group of proteins. Whereas structural proteins besides belonging to same structural group of proteins (capsid, structural, envelope), they also perform functions like nuclear receptor, antibiotic resistance, RNA-binding, DNA-binding, magnesium-binding, isomerase (intra-molecular), oxidoreductase and participate in type II (general) secretory pathway (IISP).     

Functionality and specific membrane localization of transport GTPases carrying C-terminal membrane anchors of synaptobrevin-like proteins.

Ras-related guanine nucleotide-binding proteins of the Ypt/Rab family fulfill a pivotal role in vesicular protein transport both in yeast and in mammalian cells. Proper functioning of these proteins involves their cycling between a GTP- and a GDP-bound state as well as their reversible association with specific membranes. Here we show that the yeast Ypt1 and Sec4 proteins, essential components of the vesicular transport machinery, allow unimpaired vesicular transport when permanently fixed to membranes by membrane-spanning domains replacing their two C-terminal cysteine residues. Membrane detachment of the GTPases therefore is not obligatory for transport vesicle docking to or fusion with an acceptor membrane. It was also found that the membrane anchors derived from different synaptobrevin-related proteins have targeting information and direct the chimeric GTPases to different cellular compartments, presumably from the endoplasmic reticulum via the secretory pathway.