BagReg: Protein inference through machine learning

Protein inference from the identified peptides is of primary importance in the shotgun proteomics. The target of protein inference is to identify whether each candidate protein is truly present in the sample. To date, many computational methods have been proposed to solve this problem. However, there is still no method that can fully utilize the information hidden in the input data.In this article, we propose a learning-based method named BagReg for protein inference. The method firstly artificially extracts five features from the input data, and then chooses each feature as the class feature to separately build models to predict the presence probabilities of proteins. Finally, the weak results from five prediction models are aggregated to obtain the final result. We test our method on six public available data sets. The experimental results show that our method is superior to the state-of-the-art protein inference algorithms.     

Isotope coded protein label quantification of serum proteins—Comparison with the label-free LC-MS and validation using the MRM approach

Protein quantification based upon mass spectrometry is gaining ground in diverse applications of biological and clinical relevance. The present article focuses on one of the most complex biological fluids - serum - and provides a novel ICPL based quantification protocol. The results are compared to a label-free (data independent alternate scanning) absolute quantification method. The validation is performed using MRM based protein quantification technique. Regarding the ICPL approach, serum samples used in this study were depleted of high abundant proteins, labeled with ICPL and fractionated according to their respective pI (3-5, 5-7 and 7-12). The samples were further subjected to tryptic digestion followed by treatment with the Glu-C enzyme. The peptides were analyzed on a 2D-nano-LC system using four different concentrations of salt injections (45, 75, 150 and 500 mM ammonium acetate). The LC system was connected on-line with the electrospray ion-trap mass spectrometer. For the label-free quantification the serum samples were depleted and digested with trypsin. A proteome-wide comparison was performed using highly reproducible LC and data independent alternate scanning in conjunction with a high mass accuracy orthogonal time-of-flight mass spectrometer. Selected proteins, found by both methods, were validated using the MRM approach. For this purpose non-depleted tryptically digested serum samples were analyzed by LC coupled with a triple-quadrupole MS. The relative protein quantification using ICPL and mass spectrometry allowed for the detection of approximately 200 proteins, whereas about 2/3 of those contained the ICPL label and could therefore be quantified. Label-free approach used no fractionation, less sample and was able to identify and quantify over 110 proteins. The identified proteins covered generally 3-4 orders of magnitude of protein concentration in human serum. Changes in relative abundance of eight proteins were validated using MRM. This study, for the first time, shows the ability of the relative protein quantification based upon ICPL and 2D-LC-MS/MS to quantify serum biomarkers. It provides two additional label-free approaches that could validate and bring additional value to the label-based results, offering a starting point for comprehensive proteomics studies aiming at revealing biomarkers of clinical relevance.     

Direct quantification of protein partitioning in oil-in-water emulsion by front-face fluorescence: avoiding the need for centrifugation

The quantification of proteins adsorbed at the oil-in-water interface is often difficult since it requires separation of fat globules from the aqueous phase that may damage the fat globule size and/or modify the interfacial composition. Front-face fluorescence spectroscopy was used to characterize the protein partitioning between the aqueous and oil phases of emulsions without separating these two phases. Different emulsions based on skim milk powder (SMP), two mono- and di-glyceride (MDG) mixtures (saturated and partially unsaturated), and three fats (hydrogenated and refined coconut oils and refined palm oil) were studied. The impact of an ageing period (24 h at 4 °C) was also investigated to typify the first step of ice cream processing. The emulsions were characterized for protein partitioning, immediately following emulsification and after ageing, using the Bradford spectrophotometric method, applied to the aqueous phase recovered after emulsion centrifugation. In parallel, the emulsions were characterized by their tryptophan emission fluorescence spectra. The area of the peaks at 333 nm, of the fourth-derivative fluorescence spectra corresponding to the amount of proteins present in the aqueous phase of emulsions, was well correlated with the Bradford measurements (r² = 0.91). This amount was also calculated from the fluorescence calibration curve obtained with SMP in solution. In conclusion, front-face fluorescence spectroscopy appeared to be a powerful and simple technique allowing the quantification of different populations of protein in an emulsified system, i.e., in the aqueous phase and loaded at the fat globule interface.     

Sensitive targeted multiple protein quantification based on elemental detection of Quantum Dots

A generic strategy based on the use of CdSe/ZnS Quantum Dots (QDs) as elemental labels for protein quantification, using immunoassays with elemental mass spectrometry (ICP-MS), detection is presented. In this strategy, streptavidin modified QDs (QDs-SA) are bioconjugated to a biotinylated secondary antibody (b-Ab₂). After a multi-technique characterization of the synthesized generic platform (QDs-SA-b-Ab₂) it was applied to the sequential quantification of five proteins (transferrin, complement C3, apolipoprotein A1, transthyretin and apolipoprotein A4) at different concentration levels in human serum samples. It is shown how this generic strategy does only require the appropriate unlabeled primary antibody for each protein to be detected. Therefore, it introduces a way out to the need for the cumbersome and specific bioconjugation of the QDs to the corresponding specific recognition antibody for every target analyte (protein). Results obtained were validated with those obtained using UV–vis spectrophotometry and commercial ELISA Kits.     

Indirect ultrasonication for protein quantification and peptide mass mapping through mass spectrometry-based techniques

We report in this work a fast protocol for protein quantification and for peptide mass mapping that rely on ¹⁸O isotopic labeling through the decoupling procedure. It is demonstrated that the purity and source of trypsin do not compromise the labeling degree and efficiency of the decoupled labeling reaction, and that the pH of the labeling reaction is a critical factor to obtain a significant ¹⁸O double labeling. We also show that the same calibration curve can be used for MALDI protein quantification during several days maintaining a reasonable accuracy, thus simplifying the handling of the quantification process. In addition we demonstrate that ¹⁸O isotopic labeling through the decoupling procedure can be successfully used to elaborate peptide mass maps. BSA was successfully quantified using the same calibration curve in different days and plasma from a freshwater fish, Cyprinus carpio, was used to elaborate the peptide mass maps.     

Prediction of protein structural class based on symmetrical recurrence quantification analysis

Protein structural class prediction for low similarity sequences is a significant challenge and one of the deeply explored subjects. This plays an important role in drug design, folding recognition of protein, functional analysis and several other biology applications. In this paper, we worked with two benchmark databases existing in the literature (1) 25PDB and (2) 1189 to apply our proposed method for predicting protein structural class. Initially, we transformed protein sequences into DNA sequences and then into binary sequences. Furthermore, we applied symmetrical recurrence quantification analysis (the new approach), where we got 8 features from each symmetry plot computation. Moreover, the machine learning algorithms such as Linear Discriminant Analysis (LDA), Random Forest (RF) and Support Vector Machine (SVM) are used. In addition, comparison was made to find the best classifier for protein structural class prediction. Results show that symmetrical recurrence quantification as feature extraction method with RF classifier outperformed existing methods with an overall accuracy of 100% without overfitting.     

Improved dynamic range of protein quantification in silver‐stained gels by modelling gel images over time

Silver staining is a commonly used protein stain to visualise proteins separated by 2‐DE. Despite this, the technique suffers from a limited dynamic range, making the simultaneous quantification of high‐ and low‐abundant proteins difficult. In this paper we take advantage of the fact that silver staining is not an end‐point stain by photographing the gels during development. This procedure provides information about the change in measured absorbance for each pixel in the protein spots on the gel. The maximum rate of change was found to be correlated with the amount of applied protein, providing a new way of estimating protein amount in 2‐DE gels. We observed an improvement in the dynamic range of silver staining by up to two orders of magnitude.     

Decision peptide-driven: A free software tool for accurate protein quantification using gel electrophoresis and matrix assisted laser desorption ionization time of flight mass spectrometry

The decision peptide-driven tool implements a software application for assisting the user in a protocol for accurate protein quantification based on the following steps: (1) protein separation through gel electrophoresis; (2) in-gel protein digestion; (3) direct and inverse ¹⁸O-labeling and (4) matrix assisted laser desorption ionization time of flight mass spectrometry, MALDI analysis. The DPD software compares the MALDI results of the direct and inverse ¹⁸O-labeling experiments and quickly identifies those peptides with paralleled loses in different sets of a typical proteomic workflow. Those peptides are used for subsequent accurate protein quantification. The interpretation of the MALDI data from direct and inverse labeling experiments is time-consuming requiring a significant amount of time to do all comparisons manually. The DPD software shortens and simplifies the searching of the peptides that must be used for quantification from a week to just some minutes. To do so, it takes as input several MALDI spectra and aids the researcher in an automatic mode (i) to compare data from direct and inverse ¹⁸O-labeling experiments, calculating the corresponding ratios to determine those peptides with paralleled losses throughout different sets of experiments; and (ii) allow to use those peptides as internal standards for subsequent accurate protein quantification using ¹⁸O-labeling. In this work the DPD software is presented and explained with the quantification of protein carbonic anhydrase.     

A novel (18)O inverse labeling-based workflow for accurate bottom-up mass spectrometry quantification of proteins separated by gel electrophoresis

In the present work we report on a novel and fast protocol for accurate bottom-up protein quantification that overcomes the drawbacks of in-gel digestion and MALDI analysis, while maintaining their benefits. It relies on the following steps: (i) gel electrophoresis separation of proteins, (ii) fast in-gel protein digestion with trypsin, (iii) (18)O-labeling through the decoupled method, (iv) quantification through selected peptides previously chosen using the (18)O inverse labeling approach and that, finally, (v) it takes advantage of software specifically developed to select the peptides that will drive the quantification of the protein in an automated mode. We have accurately quantified the following six proteins: glycogen phosphorylase, BSA, ovalbumin, carbonic anhydrase, trypsin inhibitor, and α-lactalbumin. As a case study we have quantified the protein vitellogenin in plasma of Cyprinus carpio exposed to high levels of estrogens. The proposed new protocol was validated against the traditional ELISA method; both were found to provide comparable results (non-parametric Mann-Whitney U-test).     

FPDock: Protein–protein docking using flower pollination algorithm

Proteins play their vital role in biological systems through interaction and complex formation with other biological molecules. Indeed, abnormalities in the interaction patterns affect the proteins’ structure and have detrimental effects on living organisms. Research in structure prediction gains its gravity as the functions of proteins depend on their structures. Protein–protein docking is one of the computational methods devised to understand the interaction between proteins. Metaheuristic algorithms are promising to use owing to the hardness of the structure prediction problem. In this paper, a variant of the Flower Pollination Algorithm (FPA) is applied to get an accurate protein–protein complex structure. The algorithm begins execution from a randomly generated initial population, which gets flourished in different isolated islands, trying to find their local optimum. The abiotic and biotic pollination applied in different generations brings diversity and intensity to the solutions. Each round of pollination applies an energy-based scoring function whose value influences the choice to accept a new solution. Analysis of final predictions based on CAPRI quality criteria shows that the proposed method has a success rate of 58% in top10 ranks, which in comparison with other methods like SwarmDock, pyDock, ZDOCK is better. Source code of the work is available at: https://github.com/Sharon1989Sunny/_FPDock_.     

A method for protein accessibility prediction based on residue types and conformational states

Prediction of protein accessibility from sequence, as prediction of protein secondary structure is an intermediate step for predicting structures and consequently functions of proteins. Most of the currently used methods are based on single residue prediction, either by statistical means or evolutionary information, and accessibility state of central residue in a window predicted. By expansion of databases of proteins with known 3D structures, we extracted information of pairwise residue types and conformational states of pairs simultaneously. For solving the problem of ambiguity in state prediction by one residue window sliding, we used dynamic programming algorithm to find the path with maximum score. The three state overall per-residue accuracy, Q₃, of this method in a Jackknife test with dataset of known proteins is more than 65% which is an improvement on results of methods based on evolutionary information.     

Protein Oligomer Engineering: A New Frontier for Studying Protein Structure,Function, and Toxicity

Prevalent in nature, protein oligomers play critical roles both physiologically and pathologically. The multimeric nature and conformational transiency of protein oligomers greatly complicate a more detailed glimpse into the molecular structure as well as function. In this minireview, the oligomers are classified and described on the basis of biological function, toxicity, and application. We also define the bottlenecks in recent oligomer studies and further review numerous frontier methods for engineering protein oligomers. Progress is being made on many fronts for a wide variety of applications, and protein grafting is highlighted as a promising and robust method for oligomer engineering. These advances collectively allow the engineering and design of stabilized oligomers that bring us one step closer to understanding their biological functions, toxicity, and a wide range of applications.     

A protein extraction method for low protein concentration solutions compatible with the proteomic analysis of rubber particles

The extraction of high‐purity proteins from the washing solution (WS) of rubber particles (also termed latex‐producing organelles) from laticifer cells in rubber tree for proteomic analysis is challenging due to the low concentration of proteins in the WS. Recent studies have revealed that proteins in the WS might play crucial roles in natural rubber biosynthesis. To further examine the involvement of these proteins in natural rubber biosynthesis, we designed an efficiency method to extract high‐purity WS proteins. We improved our current borax and phenol‐based method by adding reextraction steps with phenol (REP) to improve the yield from low protein concentration samples. With this new method, we extracted WS proteins that were suitable for proteomics. Indeed, compared to the original borax and phenol‐based method, the REP method improved both the quality and quantity of isolated proteins. By repeatedly extracting from low protein concentration solutions using the same small amount of phenol, the REP method yielded enough protein of sufficiently high‐quality from starting samples containing less than 0.02 mg of proteins per milliliter. This method was successfully applied to extract the rubber particle proteins from the WS of natural rubber latex samples. The REP‐extracted WS proteins were resolved by 2DE, and 28 proteins were positively identified by MS. This method has the potential to become widely used for the extraction of proteins from low protein concentration solutions for proteomic analysis.     

A method combining SPITC and 18O labeling for simultaneous protein identification and relative quantification

The relative quantification and identification of proteins by matrix‐assisted laser desorption ionization time‐of‐flight MS is very important in /MS is very important in protein research and is usually conducted separately. Chemical N‐terminal derivatization with 4‐sulphophenyl isothiocyanate facilitates de novo sequencing analysis and accurate protein identification, while ¹⁸O labeling is simple, specific and widely applicable among the isotopic labeling methods used for relative quantification. In the present study, a method combining 4‐sulphophenyl isothiocyanate derivatization with ¹⁸O isotopic labeling was established to identify and quantify proteins simultaneously in one experiment. Reaction conditions were first optimized using a standard peptide (fibrin peptide) and tryptic peptides from the model protein (bovine serum albumin). Under the optimized conditions, these two independent labeling steps show good compatibility, and the linear relativity of quantification within the ten times dynamic range was stable as revealed by correlation coefficient analysis (R² value = 0.998); moreover, precursor peaks in MS/MS spectrum could provide accurate quantitative information, which is usually acquired from MS spectrum, enabling protein identification and quantification in a single MS/MS spectrum. Next, this method was applied to native peptides isolated from spider venoms. As expected, the de novo sequencing results of each peptide matched with the known sequence precisely, and the measured quantitative ratio of each peptide corresponded well with the theoretical ratio. Finally, complex protein mixtures of spider venoms from male and female species with unknown genome information were analyzed. Differentially expressed proteins were successfully identified, and their quantitative information was also accessed. Taken together, this protein identification and quantification method is simple, reliable and efficient, which has a good potential in the exploration of peptides/proteins from species with unknown genome. Copyright © 2014 John Wiley & Sons, Ltd.     

High‐capacity protein A affinity chromatography for the fast quantification of antibodies: Two‐wavelength detection expands linear range

The high‐throughput analysis of antibodies from processes can be enhanced when the linear range is expanded and sample preparation is kept to a minimum. We developed a fast chromatography method based on a hexameric variant of staphylococcal protein A immobilized on Toyopearl matrix, TSK 5 PW using two wavelengths. A protocol with 5 min runtime and a single‐wavelength detection at 280 nm yielded an upper limit of quantification of 2.10 mg/mL and a lower limit of quantification of 0.06 mg/mL. The optimized method with a runtime of 2 min and two‐wavelength detection at 280 and 300 nm allowed us to span a valid concentration range of 0.01–5.20 mg/mL using two calibration curves. Sample selectivity was tested using mock supernatant mixed with antibody concentrations of 0.1–2.1 mg/mL, sample stability in the autosampler was shown for at least 24 h. We also tested the capabilities of the method to determine purity of an antibody sample by calculating the ratio of peak area of elution to peak area of flow‐through, which correlated well with the expected purity. The method will be very useful for process development and in‐process control, spanning concentrations from seed fermentation to harvest and purification.     

A protein extraction method compatible with proteomic analysis for the euhalophyte Salicornia europaea

Protein extraction from plants like the halophyte Salicornia europaea has been problematic using standard protocols due to high concentrations of salt ions in their cells. We have developed an improved method for protein extraction from S. europaea, which allowed us to remove interfering compounds and salt ions by including the chemicals borax, polyvinylpolypyrrolidone, and phenol. The comparative study of this method with several other protocols using NaCl-treated S. europaea shoots demonstrated that this method gave the best distinction of proteins on 2-DE gels. This protocol had a wide range of applications as high yields and good distinction of 1-DE gels for proteins isolated from twelve other plants were rendered. In addition, we reported results of 2-DE using the recalcitrant tissue of the S. europaea roots. We also demonstrated that this protocol is compatible with proteomic analysis as eight specific proteins generated by this method have been identified by MS. In conclusion, our newly developed protein extraction protocol is expected to have excellent applications in proteomic studies of halophytes.     

A benchmark of optimally folded protein structures using integer programming and the 3D-HP-SC model

The Protein Structure Prediction (PSP) problem comprises, among other issues, forecasting the three-dimensional native structure of proteins using only their primary structure information. Most computational studies in this area use synthetic data instead of real biological data. However, the closer to the real-world, the more the impact of results and their applicability. This work presents 17 real protein sequences extracted from the Protein Data Bank for a benchmark to the PSP problem using the tri-dimensional Hydrophobic-Polar with Side-Chains model (3D-HP-SC). The native structure of these proteins was found by maximizing the number of hydrophobic contacts between the side-chains of amino acids. The problem was treated as an optimization problem and solved by means of an Integer Programming approach. Although the method optimally solves the problem, the processing time has an exponential trend. Therefore, due to computational limitations, the method is a proof-of-concept and it is not applicable to large sequences. For unknown sequences, an upper bound of the number of hydrophobic contacts (using this model) can be found, due to a linear relationship with the number of hydrophobic residues. The comparison between the predicted and the biological structures showed that the highest similarity between them was found with distance thresholds around 5.2–8.2 Å. Both the dataset and the programs developed will be freely available to foster further research in the area.     

Quantification of immunoglobulin G and characterization of process related impurities using coupled Protein A and size exclusion high performance liquid chromatography

The present work describes two HPLC-UV methods for multi-protein quantification using (i) only a Protein A sensor cartridge (Protein A HPLC) and (ii) the same Protein A cartridge in combination with a size exclusion HPLC column (PSEC-HPLC). The possibility to simultaneously quantify immunoglobulin G (IgG) besides a non-binding protein such as bovine serum albumin (BSA) increases the applicability of Protein A HPLC. Its most pronounced feature is its independence of the buffer system, pH-value and salt content of the investigated sample solvent, which includes cell media. A comparison with the state-of-the-art, the photometrical Bradford method, shows that Protein A HPLC is as sensitive as Bradford, but that it comes with an extended linear range of 4 orders of magnitude, ranging from 0.15 [μg abs] to 1 [mg abs] absolute injected protein amount. The applicability of the PSEC-HPLC method is demonstrated for the analysis of real cell culture feed samples. While Protein A binds IgG, the SEC-column distributes the feed impurities by their molecular weight. The peak area ratios of IgG and the feed impurities of interest are then plotted against the collected sample fraction. These Protein A-Size-Exclusion-Chromatographic diagrams (PSEC-plot) combine the performance information of feed impurities and IgG in a single plot. Further it is shown that both methods are suitable for the performance evaluation of antibody purification media using static as well as dynamic binding experiments performed on DEAE-Fractogel and Capto Adhere. The investigated test samples were “mock” protein solutions with increasing complexity ranging from simple PBS buffer to serum free cell media and “real” cell culture feed solutions.     

Protein Folding: A Perspective from Theory and Experiment

The mechanism of protein folding (represented schematically below) is one of the most fascinating problems in the field of chemical reactions. This review presents the progess made recently in understanding key elements of this reaction and describes a solution to the often quoted Levinthal Paradox.     

Characterization of quinoa seed proteome combining different protein precipitation techniques: Improvement of knowledge of nonmodel plant proteomics

A shotgun proteomics approach was used to characterize the quinoa seed proteome. To obtain comprehensive proteomic data from quinoa seeds three different precipitation procedures were employed: MeOH/CHCl₃/double‐distilled H₂O, acetone either alone or with trichloroacetic acid; the isolated proteins were then in‐solution digested and the resulting peptides were analyzed by nano‐liquid chromatography coupled to tandem mass spectrometry. However, since quinoa is a nonmodel plant species, only a few protein sequences are included in the most widely known protein sequence databases. To improve the data reliability a UniProt subdatabase, containing only proteins of Caryophillales order, was used. A total of 352 proteins were identified and evaluated both from a qualitative and quantitative point of view. This combined approach is certainly useful to increase the final number of identifications, but no particular class of proteins was extracted and identified in spite of the different chemistries and the different precipitation protocols. However, with respect to the other two procedures, from the relative quantitative analysis, based on the number of spectral counts, the trichloroacetic acid/acetone protocol was the best procedure for sample handling and quantitative protein extraction. This study could pave the way to further high‐throughput studies on Chenopodium Quinoa.