Subfamily specific conservation profiles for proteins based on n-gram patterns

Time series computational prediction of vaccines for influenza A H3N2 with recurrent neural networks

Influenza viruses are persistently threatening public health, causing annual epidemics and sporadic pandemics due to rapid viral evolution. Vaccines are used to prevent influenza infections but the composition of the influenza vaccines have to be updated regularly to ensure its efficacy. Computational tools and analyses have become increasingly important in guiding the process of vaccine selection. By constructing time-series training samples with splittings and embeddings, we develop a computational method for predicting suitable strains as the recommendation of the influenza vaccines using recurrent neural networks (RNNs). The Encoder-decoder architecture of RNN model enables us to perform sequence-to-sequence prediction. We employ this model to predict the prevalent sequence of the H3N2 viruses sampled from 2006 to 2017. The identity between our predicted sequence and recommended vaccines is greater than 98% and the Pepitope<0.2 indicates their antigenic similarity. The multi-step vaccine prediction further demonstrates the robustness of our method which achieves comparable results in contrast to single step prediction. The results show significant matches of the recommended vaccine strains to the circulating strains. We believe it would facilitate the process of vaccine selection and surveillance of seasonal influenza epidemics.

Machine learning predicts nucleosome binding modes of transcription factors

Background

Most transcription factors (TFs) compete with nucleosomes to gain access to their cognate binding sites. Recent studies have identified several TF-nucleosome interaction modes including end binding (EB), oriented binding, periodic binding, dyad binding, groove binding, and gyre spanning. However, there are substantial experimental challenges in measuring nucleosome binding modes for thousands of TFs in different species.

Results

We present a computational prediction of the binding modes based on TF protein sequences. With a nested cross-validation procedure, our model outperforms several fine-tuned off-the-shelf machine learning (ML) methods in the multi-label classification task. Our binary classifier for the EB mode performs better than these ML methods with the area under precision-recall curve achieving 75%. The end preference of most TFs is consistent with low nucleosome occupancy around their binding site in GM12878 cells. The nucleosome occupancy data is used as an alternative dataset to confirm the superiority of our EB classifier.

Conclusions

We develop the first ML-based approach for efficient and comprehensive analysis of nucleosome binding modes of TFs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04093-9.

Machine learning can be used to distinguish protein families and generate new proteins belonging to those families

Proteins are classified into families based on evolutionary relationships and common structure-function characteristics. Availability of large data sets of gene-derived protein sequences drives this classification. Sequence space is exponentially large, making it difficult to characterize family differences. In this work, we show that Machine Learning (ML) methods can be trained to distinguish between protein families. A number of supervised ML algorithms are explored to this end. The most accurate is a Long Short Term Memory (LSTM) classification method that accounts for the sequence context of the amino acids. Sequences for a number of protein families where there are sufficient data to be used in ML are studied. By splitting the data into training and testing sets, we find that this LSTM classifier can be trained to successfully classify the test sequences for all pairs of the families. Also investigated is whether the addition of structural information increases the accuracy of the binary comparisons. It does, but because there is much less available structural than sequence information, the quality of the training degrades. Another variety of LSTM, LSTM_wordGen, a context-dependent word generation algorithm, is used to generate new protein sequences based on seed sequences for the families considered here. Using the original sequences as training data and the generated sequences as test data, the LSTM classification method classifies the generated sequences almost as accurately as the true family members do. Thus, in principle, we have generated new members of these protein families.

A New Method for Recognizing Cytokines Based on Feature Combination and a Support Vector Machine Classifier

Research on cytokine recognition is of great significance in the medical field due to the fact cytokines benefit the diagnosis and treatment of diseases, but the current methods for cytokine recognition have many shortcomings, such as low sensitivity and low F-score. Therefore, this paper proposes a new method on the basis of feature combination. The features are extracted from compositions of amino acids, physicochemical properties, secondary structures, and evolutionary information. The classifier used in this paper is SVM. Experiments show that our method is better than other methods in terms of accuracy, sensitivity, specificity, F-score and Matthew's correlation coefficient.

Continuous Distributed Representation of Biological Sequences for Deep Proteomics and Genomics

We introduce a new representation and feature extraction method for biological sequences. Named bio-vectors (BioVec) to refer to biological sequences in general with protein-vectors (ProtVec) for proteins (amino-acid sequences) and gene-vectors (GeneVec) for gene sequences, this representation can be widely used in applications of deep learning in proteomics and genomics. In the present paper, we focus on protein-vectors that can be utilized in a wide array of bioinformatics investigations such as family classification, protein visualization, structure prediction, disordered protein identification, and protein-protein interaction prediction. In this method, we adopt artificial neural network approaches and represent a protein sequence with a single dense n-dimensional vector. To evaluate this method, we apply it in classification of 324,018 protein sequences obtained from Swiss-Prot belonging to 7,027 protein families, where an average family classification accuracy of 93%±0.06% is obtained, outperforming existing family classification methods. In addition, we use ProtVec representation to predict disordered proteins from structured proteins. Two databases of disordered sequences are used: the DisProt database as well as a database featuring the disordered regions of nucleoporins rich with phenylalanine-glycine repeats (FG-Nups). Using support vector machine classifiers, FG-Nup sequences are distinguished from structured protein sequences found in Protein Data Bank (PDB) with a 99.8% accuracy, and unstructured DisProt sequences are differentiated from structured DisProt sequences with 100.0% accuracy. These results indicate that by only providing sequence data for various proteins into this model, accurate information about protein structure can be determined. Importantly, this model needs to be trained only once and can then be applied to extract a comprehensive set of information regarding proteins of interest. Moreover, this representation can be considered as pre-training for various applications of deep learning in bioinformatics. The related data is available at Life Language Processing Website: http://llp.berkeley.edu and Harvard Dataverse: http://dx.doi.org/10.7910/DVN/JMFHTN.

An analysis approach to identify specific functional sites in orthologous proteins using sequence and structural information: application to neuroserpin reveals regions that differentially regulate inhibitory activity

The analysis of sequence conservation is commonly used to predict functionally important sites in proteins. We have developed an approach that first identifies highly conserved sites in a set of orthologous sequences using a weighted substitution-matrix-based conservation score and then filters these conserved sites based on the pattern of conservation present in a wider alignment of sequences from the same family and structural information to identify surface-exposed sites. This allows us to detect specific functional sites in the target protein and exclude regions that are likely to be generally important for the structure or function of the wider protein family. We applied our method to two members of the serpin family of serine protease inhibitors. We first confirmed that our method successfully detected the known heparin binding site in antithrombin while excluding residues known to be generally important in the serpin family. We next applied our sequence analysis approach to neuroserpin and used our results to guide site-directed polyalanine mutagenesis experiments. The majority of the mutant neuroserpin proteins were found to fold correctly and could still form inhibitory complexes with tissue plasminogen activator (tPA). Kinetic analysis of tPA inhibition, however, revealed altered inhibitory kinetics in several of the mutant proteins, with some mutants showing decreased association with tPA and others showing more rapid dissociation of the covalent complex. Altogether, these results confirm that our sequence analysis approach is a useful tool that can be used to guide mutagenesis experiments for the detection of specific functional sites in proteins.

Mining for class-specific motifs in protein sequence classification

BACKGROUND

In protein sequence classification, identification of the sequence motifs or n-grams that can precisely discriminate between classes is a more interesting scientific question than the classification itself. A number of classification methods aim at accurate classification but fail to explain which sequence features indeed contribute to the accuracy. We hypothesize that sequences in lower denominations (n-grams) can be used to explore the sequence landscape and to identify class-specific motifs that discriminate between classes during classification. Discriminative n-grams are short peptide sequences that are highly frequent in one class but are either minimally present or absent in other classes. In this study, we present a new substitution-based scoring function for identifying discriminative n-grams that are highly specific to a class.

RESULTS

We present a scoring function based on discriminative n-grams that can effectively discriminate between classes. The scoring function, initially, harvests the entire set of 4- to 8-grams from the protein sequences of different classes in the dataset. Similar n-grams of the same size are combined to form new n-grams, where the similarity is defined by positive amino acid substitution scores in the BLOSUM62 matrix. Substitution has resulted in a large increase in the number of discriminatory n-grams harvested. Due to the unbalanced nature of the dataset, the frequencies of the n-grams are normalized using a dampening factor, which gives more weightage to the n-grams that appear in fewer classes and vice-versa. After the n-grams are normalized, the scoring function identifies discriminative 4- to 8-grams for each class that are frequent enough to be above a selection threshold. By mapping these discriminative n-grams back to the protein sequences, we obtained contiguous n-grams that represent short class-specific motifs in protein sequences. Our method fared well compared to an existing motif finding method known as Wordspy. We have validated our enriched set of class-specific motifs against the functionally important motifs obtained from the NLSdb, Prosite and ELM databases. We demonstrate that this method is very generic; thus can be widely applied to detect class-specific motifs in many protein sequence classification tasks.

CONCLUSION

The proposed scoring function and methodology is able to identify class-specific motifs using discriminative n-grams derived from the protein sequences. The implementation of amino acid substitution scores for similarity detection, and the dampening factor to normalize the unbalanced datasets have significant effect on the performance of the scoring function. Our multipronged validation tests demonstrate that this method can detect class-specific motifs from a wide variety of protein sequence classes with a potential application to detecting proteome-specific motifs of different organisms.

Word decoding of protein amino Acid sequences with availability analysis: a linguistic approach

The amino acid sequences of proteins determine their three-dimensional structures and functions. However, how sequence information is related to structures and functions is still enigmatic. In this study, we show that at least a part of the sequence information can be extracted by treating amino acid sequences of proteins as a collection of English words, based on a working hypothesis that amino acid sequences of proteins are composed of short constituent amino acid sequences (SCSs) or "words". We first confirmed that the English language highly likely follows Zipf's law, a special case of power law. We found that the rank-frequency plot of SCSs in proteins exhibits a similar distribution when low-rank tails are excluded. In comparison with natural English and "compressed" English without spaces between words, amino acid sequences of proteins show larger linear ranges and smaller exponents with heavier low-rank tails, demonstrating that the SCS distribution in proteins is largely scale-free. A distribution pattern of SCSs in proteins is similar among species, but species-specific features are also present. Based on the availability scores of SCSs, we found that sequence motifs are enriched in high-availability sites (i.e., "key words") and vice versa. In fact, the highest availability peak within a given protein sequence often directly corresponds to a sequence motif. The amino acid composition of high-availability sites within motifs is different from that of entire motifs and all protein sequences, suggesting the possible functional importance of specific SCSs and their compositional amino acids within motifs. We anticipate that our availability-based word decoding approach is complementary to sequence alignment approaches in predicting functionally important sites of unknown proteins from their amino acid sequences.

A singular value decomposition approach for improved taxonomic classification of biological sequences

Background

Singular value decomposition (SVD) is a powerful technique for information retrieval; it helps uncover relationships between elements that are not prima facie related. SVD was initially developed to reduce the time needed for information retrieval and analysis of very large data sets in the complex internet environment. Since information retrieval from large-scale genome and proteome data sets has a similar level of complexity, SVD-based methods could also facilitate data analysis in this research area.

Results

We found that SVD applied to amino acid sequences demonstrates relationships and provides a basis for producing clusters and cladograms, demonstrating evolutionary relatedness of species that correlates well with Linnaean taxonomy. The choice of a reasonable number of singular values is crucial for SVD-based studies. We found that fewer singular values are needed to produce biologically significant clusters when SVD is employed. Subsequently, we developed a method to determine the lowest number of singular values and fewest clusters needed to guarantee biological significance; this system was developed and validated by comparison with Linnaean taxonomic classification.

Conclusions

By using SVD, we can reduce uncertainty concerning the appropriate rank value necessary to perform accurate information retrieval analyses. In tests, clusters that we developed with SVD perfectly matched what was expected based on Linnaean taxonomy.