Prediction of optimal pH and temperature of cellulases using neural network

Cellulase is an important enzyme widely used in various industries, and now in fermentation of biomass into biofuels. Enzymatic function of cellulase is closely related to pH, temperature, substrate concentration, etc. For newly found cellulase, it would be more cost-effective to predict its optimal pH and temperature before conducting the costly experiments. In this study, we used a 20-2 feedforward backpropagation neural network to build the relationship between information obtained from primary structure of cellulase with optimal pH and temperature to predict the optimal pH and temperature in cellulases. The results show that the amino-acid distribution probability representing the primary structure of cellulase can predict both optimal pH and temperature, whereas various properties of amino acids related to the primary structure cannot do so.     

Prediction of siRNA knockdown efficiency using artificial neural network models

Selective knockdown of gene expression by short interference RNAs (siRNAs) has allowed rapid validation of gene functions and made possible a high throughput, genome scale approach to interrogate gene function. However, randomly designed siRNAs display different knockdown efficiencies of target genes. Hence, various prediction algorithms based on siRNA functionality have recently been constructed to increase the likelihood of selecting effective siRNAs, thereby reducing the experimental cost. Toward this end, we have trained three Back-propagation and Bayesian neural network models, previously not used in this context, to predict the knockdown efficiencies of 180 experimentally verified siRNAs on their corresponding target genes. Using our input coding based primarily on RNA structure thermodynamic parameters and cross-validation method, we showed that our neural network models outperformed most other methods and are comparable to the best predicting algorithm thus far published. Furthermore, our neural network models correctly classified 74% of all siRNAs into different efficiency categories; with a correlation coefficient of 0.43 and receiver operating characteristic curve score of 0.78, thus highlighting the potential utility of this method to complement other existing siRNA classification and prediction schemes.     

基于卷积神经网络的细菌转录终止子预测

在遗传学中,终止子是位于poly(A)位点下游、长度在数百碱基以内、包含多个回文序列、具有终止转录功能的DNA结构域,其主要作用是使转录终止。在原核生物基因组中有两类转录终止子,即Rho-dependent因子和Rho-independent因子。在本项研究中,提出了一种新的预测模型(TermCNN)来快速准确地识别细菌转录终止子。该模型将具有代表性的6-mer特征子集(2 537个特征)和电子—离子相互作用伪电位(EIIP)作为输入向量,利用卷积神经网络(CNN)构建预测模型。五折交叉验证和独立测试的结果表明该模型优于最新的预测模型iTerm-PseKNC。值得注意的是,该模型在跨物种试验中具有明显的优势。它可以高度精确地预测大肠杆菌(E. coli)和枯草芽孢杆菌(B. subtilis)的转录终止子。     

Prediction of beta-turns in proteins from multiple alignment using neural network

A neural network-based method has been developed for the prediction of beta-turns in proteins by using multiple sequence alignment. Two feed-forward back-propagation networks with a single hidden layer are used where the first-sequence structure network is trained with the multiple sequence alignment in the form of PSI-BLAST-generated position-specific scoring matrices. The initial predictions from the first network and PSIPRED-predicted secondary structure are used as input to the second structure-structure network to refine the predictions obtained from the first net. A significant improvement in prediction accuracy has been achieved by using evolutionary information contained in the multiple sequence alignment. The final network yields an overall prediction accuracy of 75.5% when tested by sevenfold cross-validation on a set of 426 nonhomologous protein chains. The corresponding Q(pred), Q(obs), and Matthews correlation coefficient values are 49.8%, 72.3%, and 0.43, respectively, and are the best among all the previously published beta-turn prediction methods. The Web server BetaTPred2 (http://www.imtech.res.in/raghava/betatpred2/) has been developed based on this approach.     

Prediction of protein secondary structure using improved two-level neural network architecture

In this paper we propose constructing an improved two-level neural network to predict protein secondary structure. Firstly, we code the whole protein composition information as the inputs to the first-level network besides the evolutionary information. Secondly, we calculate the reliability score for each residue position based on the output of the first-level network, and the role of the second-level network is to take full advantage of the residues with a higher reliability score to impact the neighboring residues with a lower one for improving the whole prediction accuracy. Thirdly, considering it is indeed a problem that the target protein can be lost in the multiple sequence alignment we propose to code single sequence into the second-level network. The experimental results show that our proposed method can efficiently improve the prediction accuracy.     

Prediction of continuous B-cell epitopes in an antigen using recurrent neural network

B-cell epitopes play a vital role in the development of peptide vaccines, in diagnosis of diseases, and also for allergy research. Experimental methods used for characterizing epitopes are time consuming and demand large resources. The availability of epitope prediction method(s) can rapidly aid experimenters in simplifying this problem. The standard feed-forward (FNN) and recurrent neural network (RNN) have been used in this study for predicting B-cell epitopes in an antigenic sequence. The networks have been trained and tested on a clean data set, which consists of 700 non-redundant B-cell epitopes obtained from Bcipep database and equal number of non-epitopes obtained randomly from Swiss-Prot database. The networks have been trained and tested at different input window length and hidden units. Maximum accuracy has been obtained using recurrent neural network (Jordan network) with a single hidden layer of 35 hidden units for window length of 16. The final network yields an overall prediction accuracy of 65.93% when tested by fivefold cross-validation. The corresponding sensitivity, specificity, and positive prediction values are 67.14, 64.71, and 65.61%, respectively. It has been observed that RNN (JE) was more successful than FNN in the prediction of B-cell epitopes. The length of the peptide is also important in the prediction of B-cell epitopes from antigenic sequences. The webserver ABCpred is freely available at www.imtech.res.in/raghava/abcpred/.     

Prediction of contact numbers of amino acid residues using a neural network regression algorithm

The profile of contact numbers of amino acid residues in proteins contains important information about the protein structure and is connected with the accessibility of residues to solvent. Here we propose a method for predicting the profile of contact numbers of residues in protein from its amino acid sequence. The method is based on regression using a neural network algorithm. The algorithm predicts two types of profiles, namely, the total number of contacts and the number of close contacts with the neighbors in the chain. The Pearson coefficient of correlation between the actual and predicted values of total contact numbers amounted to 0.526–0.703. As for the number of close contacts, this coefficient was higher (0.662–0.743) for all the considered threshold contact distances (6, 8, 10, and 12 Å). The program for prediction of contact numbers CONNP is available at http://wwwmgs2.bionet.nsc.ru/reloaded.     

Prediction of contact numbers of amino acid residues using a neural network regression algorithm

The profile of contact numbers of amino acid residues in proteins contains important information about the protein structure and is connected with the accessibility of residues to solvent. Here we propose a method for predicting the profile of contact numbers of residues in protein from its amino acid sequence. The method is based on regression using a neural network algorithm. The algorithm predicts two types of profiles, namely, the total number of contacts and the number of close contacts with the neighbors in the chain. The Pearson coefficient of correlation between the actual and predicted values of total contact numbers amounted to 0.526–0.703. As for the number of close contacts, this coefficient was higher (0.662–0.743) for all the considered threshold contact distances (6, 8, 10, and 12 Å). The program for prediction of contact numbers CONNP is available at http://wwwmgs2.bionet.nsc.ru/reloaded.

     

Prediction of MHC binding peptide using Gibbs motif sampler,weight matrix and artificial neural network

The identification of MHC restricted epitopes is an important goal in peptide based vaccine and diagnostic development. As wet lab experiments for identification of MHC binding peptide are expensive and time consuming, in silico tools have been developed as fast alternatives, however with low performance. In the present study, we used IEDB training and blind validation datasets for the prediction of peptide binding to fourteen human MHC class I and II molecules using Gibbs motif sampler, weight matrix and artificial neural network methods. As compare to MHC class I predictor based on sequence weighting (Aroc=0.95 and CC=0.56) and artificial neural network (Aroc=0.73 and CC=0.25), MHC class II predictor based on Gibbs sampler did not perform well (Aroc=0.62 and CC=0.19). The predictive accuracy of Gibbs motif sampler in identifying the 9-mer cores of a binding peptide to DRB1 alleles are also limited (40¢), however above the random prediction (14¢). Therefore, the size of dataset (training and validation) and the correct identification of the binding core are the two main factors limiting the performance of MHC class-II binding peptide prediction. Overall, these data suggest that there is substantial room to improve the quality of the core predictions using novel approaches that capture distinct features of MHC-peptide interactions than the current approaches.     

Prediction of the coupling specificity of GPCRs to four families of G-proteins using hidden Markov models and artificial neural networks

MOTIVATION: G-protein coupled receptors are a major class of eukaryotic cell-surface receptors. A very important aspect of their function is the specific interaction (coupling) with members of four G-protein families. A single GPCR may interact with members of more than one G-protein families (promiscuous coupling). To date all published methods that predict the coupling specificity of GPCRs are restricted to three main coupling groups G(i/o), G(q/11) and G(s), not including G(12/13)-coupled or other promiscuous receptors. RESULTS: We present a method that combines hidden Markov models and a feed-forward artificial neural network to overcome these limitations, while producing the most accurate predictions currently available. Using an up-to-date curated dataset, our method yields a 94% correct classification rate in a 5-fold cross-validation test. The method predicts also promiscuous coupling preferences, including coupling to G(12/13), whereas unlike other methods avoids overpredictions (false positives) when non-GPCR sequences are encountered. AVAILABILITY: A webserver for academic users is available at http://bioinformatics.biol.uoa.gr/PRED-COUPLE2     

Natural base-pairing interactions between 5' splice site and branch site sequences affect mammalian 5' splice site selection.

In the murine gene encoding the neuronal cell adhesion molecule (NCAM), the integrity of the 5' splice site of exon 18 (E18) is essential for regulation of alternative splicing. To further study the contribution of 5' splice site sequences, we used a simple NCAM pre-mRNA containing a portion of E18 fused to E19 and separated by a shortened intron. This RNA is spliced in vitro to produce five sets of lariat intermediates and products, the most abundant set displaying aberrant migration in acrylamide/urea gels. Base pairing interactions between positions +5 and +8 of the intron and positions -3 and -6 from the branch point were responsible for the faster migration of this set of lariat molecules. To test whether the duplex structure forms earlier and contributes to 5' splice site selection, we used NCAM substrates carrying the 5' splice sites of E17 and E18 in competition for the 3' splice site of E19. Mutations upstream of the major branch site improve E18/E19 splicing in NIH3T3 extracts, whereas compensatory mutations at positions +7 and +8 neutralize the effect of branch site mutations and curtail E18/E19 splicing. Our data suggest that duplex formation occurs early and interferes with the assembly of complexes initiated on the 5' splice site of NCAM E18. This novel type of intron interaction may exist in the introns of other mammalian pre-mRNAs.     

PCR differentiation of commercial yeast strains using intron splice site primers.

The increased use of pure starter cultures in the wine industry has made it necessary to develop a rapid and simple identification system for yeast strains. A method based upon the PCR using oligonucleotide primers that are complementary to intron splice sites has been developed. Since most introns are not essential for gene function, introns have evolved with minimal constraint. By targeting these highly variable sequences, the PCR has proved to be very effective in uncovering polymorphisms in commercial yeast strains. The speed of the method and the ability to analyze many samples in a single day permit the monitoring of specific yeast strains during fermentations. Furthermore, the simplicity of the technique, which does not require the isolation of DNA, makes it accessible to industrial laboratories that have limited molecular expertise and resources.     

Prediction of removal efficiency of Lanaset Red G on walnut husk using artificial neural network model

An artificial neural network (ANN) model was used to predict removal efficiency of Lanaset Red (LR) G on walnut husk (WH). This adsorbent was characterized by FTIR-ATR. Effects of particle size, adsorbent dose, initial pH value, dye concentration, and contact time were investigated to optimize sorption process. Operating variables were used as the inputs to the constructed neural network to predict the dye uptake at any time as an output. Commonly used pseudo second-order model was fitted to the experimental data to compare with ANN model. According to error analyses and determination of coefficients, ANN was the more appropriate model to describe this sorption process. Results of ANN indicated that pH was the most efficient parameter (43%), followed by initial dye concentration (40%) for sorption of LR G on WH.     

Predictive glycoengineering of biosimilars using a Markov chain glycosylation model

Biosimilar drugs must closely resemble the pharmacological attributes of innovator products to ensure safety and efficacy to obtain regulatory approval. Glycosylation is one critical quality attribute that must be matched, but it is inherently difficult to control due to the complexity of its biogenesis. This usually implies that costly and time‐consuming experimentation is required for clone identification and optimization of biosimilar glycosylation. Here, a computational method that utilizes a Markov model of glycosylation to predict optimal glycoengineering strategies to obtain a specific glycosylation profile with desired properties is described. The approach uses a genetic algorithm to find the required quantities to perturb glycosylation reaction rates that lead to the best possible match with a given glycosylation profile. Furthermore, the approach can be used to identify cell lines and clones that will require minimal intervention while achieving a glycoprofile that is most similar to the desired profile. Thus, this approach can facilitate biosimilar design by providing computational glycoengineering guidelines that can be generated with a minimal time and cost.     

Image segmentation using a neural network

An object extraction problem based on the Gibbs Random Field model is discussed. The Maximum a'posteriori probability (MAP) estimate of a scene based on a noise-corrupted realization is found to be computationally exponential in nature. A neural network, which is a modified version of that of Hopfield, is suggested for solving the problem. A single neuron is assigned to every pixel. Each neuron is supposed to be connected only to all of its nearest neighbours. The energy function of the network is designed in such a way that its minimum value corresponds to the MAP estimate of the scene. The dynamics of the network are described. A possible hardware realization of a neuron is also suggested. The technique is implemented on a set of noisy images and found to be highly robust and immune to noise.