Mutation trend of hemagglutinin of influenza A virus: a review from a computational mutation viewpoint

Forecasting influenza hemagglutinin mutations through the lens of anomaly detection

The influenza virus hemagglutinin is an important part of the virus attachment to the host cells. The hemagglutinin proteins are one of the genetic regions of the virus with a high potential for mutations. Due to the importance of predicting mutations in producing effective and low-cost vaccines, solutions that attempt to approach this problem have recently gained significant attention. A historical record of mutations has been used to train predictive models in such solutions. However, the imbalance between mutations and preserved proteins is a big challenge for the development of such models that need to be addressed. Here, we propose to tackle this challenge through anomaly detection (AD). AD is a well-established field in Machine Learning (ML) that tries to distinguish unseen anomalies from normal patterns using only normal training samples. By considering mutations as anomalous behavior, we could benefit existing rich solutions in this field that have emerged recently. Such methods also fit the problem setup of extreme imbalance between the number of unmutated vs. mutated training samples. Motivated by this formulation, our method tries to find a compact representation for unmutated samples while forcing anomalies to be separated from the normal ones. This helps the model to learn a shared unique representation between normal training samples as much as possible, which improves the discernibility and detectability of mutated samples from the unmutated ones at the test time. We conduct a large number of experiments on four publicly available datasets, consisting of three different hemagglutinin protein datasets, and one SARS-CoV-2 dataset, and show the effectiveness of our method through different standard criteria.

Plasmon Nanocomposite-Enhanced Optical and Electrochemical Signals for Sensitive Virus Detection

The social impact of virus spread is immeasurable. Vaccine prophylaxes take considerable time to develop because clinical trials are required. The best initial response to an emerging virus is establishing a virus detection technology adapted by simply preparing virus-specific antibodies. A virus detection system that detects two signals from one analyte has been developed to detect the target virus more sensitively and reliably. Plasmon regions on the surface of nanoparticles are effective in enhancing optical and electrochemical signals. Thus, CdSeTeS quantum dots (QDs) have been used as optical and electrochemical signal-generating materials. In contrast, gold nanoparticle-magnetic nanoparticle-carbon nanotube (AuNP-MNP-CNT) nanocomposites are used for the magnetic separation of the virus from interferences and for signal enhancement. In the presence of the target virus, the QDs optically show a virus concentration-dependent fluorescence enhancement effect due to the localized surface plasmon resonance (LSPR) of AuNPs. Regarding the electrochemical signal, Cd ions eluted by acid degradation of the QDs in solution show a virus concentration-dependent increase in the current peak on an electrode whose electrochemical properties are improved by the deposition of these nanocomposites. Both nanomaterials are conjugated with antibodies specific to influenza virus A (IFV/A), binding this target in a sandwich structure. We are successfully detecting the virus from these two signals during actual virus detection, even when the virus particles are in a human serum matrix. The limit of detection is 2.16 fg/mL for optical detection and 13.66 fg/mL for electrochemical detection.

Serological and molecular prevalence of swine influenza virus on farms in northwestern Mexico

The aim of this study was to provide an overview of the epidemiological status of swine influenza viruses in pigs from northwestern Mexico in 2008-2009. A serological and molecular survey was conducted in 150 pigs from 15 commercial farms in Sonora, Mexico (northwestern region of Mexico). The serological data showed that 55% of the sera were positive for the H1N1 subtype, 59% for the H3N2 subtype, and 38% for both subtypes. Overall, 16.6% (25/150) of the samples were positive for type A influenza by qRT-PCR. The phylogenetic analysis of the H1 viruses circulating in northwestern Mexico were grouped into cluster α, from five other clusters previously described. The influenza virus H1 circulating in northwestern Mexico showed 97-100% identity at the nucleotide level among them, 89% identity with other North American strains, 88% with strains from central Mexico, and 85% with the pandemic A/H1N1p2009 virus. Meanwhile, a closer relationship with some influenza viruses from North America (97% nucleotide identity) was found for H3 subtype. In conclusion, our results demonstrated a high circulation of strains similar to those observed in the North American linage among commercial farms in northwestern Mexico, involving of a different lineage virus different to the influenza pandemic of 2009.

Possibility of cross-species/subtype reassortments in influenza A viruses: an analysis of nonstructural protein variations

The reassortment of genetic segments from different host species and from different subtypes of influenza A viruses occurs frequently, which may generate new strains causing flu epidemic or pandemic. However, the underlined mechanisms of reassortment were less addressed from the viewpoint of protein variations. Recently, we used the amino-acid pair predictability as an indicator to convert eight types of influenza A virus proteins into predictable portion of amino-acid pairs, and then applied the models I and II ANOVA to estimate their differences in terms of subtypes and host species. In order to get a full picture, 2729 and 1063 non-structural 1 and 2 proteins of influenza A viruses were analyzed in this study. The results are consistent with those obtained from hemagglutinin, neuraminidase, nucleoprotein, polymerase acidic protein, polymerase basic proteins 1 and 2, and matrix proteins 1 and 2, indicating that inter-species/subtypes variations are smaller than intra-species/subtype ones. Our findings provide statistical evidence that can partially explains why cross-subtype mutation and cross-species infection easily occur during co-infecting of different strains.

Prediction of optimal pH in hydrolytic reaction of beta-glucosidase

This is the continuation of our studies to use very basic information on enzyme to predict optimal reaction parameters in enzymatic reactions because the gap between available enzyme sequences and their available reaction parameters is widening. In this study, 23 features selected from 540 plus features of individual amino acid as well as a feature combined whole protein information were screened as independents in a 20-1 feedforward backpropagation neural network for predicting optimal pH in beta-glucosidase's hydrolytic reaction because this enzyme drew attention recently due to its role in biofuel industry. The results show that 11 features can be used as independents for the prediction, while the feature of amino acid distribution probability works better than the rest independents for the prediction. Our study paves a way to predict the optimal reaction parameters of enzymes based on the amino acid features of enzyme sequences.

Predicting Km values of beta-glucosidases using cellobiose as substrate

The Michaelis-Menten constant Km is a very important parameter to relate enzyme with its substrate in enzymatic reaction. Although Km can be experimentally determined, the Km values are not easily available in literature. With rapid increase of newly designed enzymes, we face the shortage of parameters related to enzymatic reactions. The beta-glucosidase is a crucial enzyme for cellulose hydrolysis and cellobiose is one of its substrates. In this study, we attempt to develop models to predict Km with cellobiose as substrates using information about primary structure of beta-glucosidase. The results show that the 20-1 feedforward backpropagation neural network using the amino-acid distribution probability as predictor works best for prediction of Km values.

Exhausted jackknife validation exemplified by prediction of temperature optimum in enzymatic reaction of cellulases

This was the continuation of our previous study along the same line with more focus on technical details because the data are usually divided into two datasets, one for model development and the other for model validation during the development of predictive model. The widely used validation method is the delete-1 jackknife validation. However, no systematical studies were conducted to determine whether the jackknife validation with different deletions works better because the number of validations with different deletions increases in a factorial fashion. Therefore it is only small dataset that can be used for such an exhausted study. Cellulase is an enzyme playing an important role in modern industry, and many parameters related to cellulase in enzymatic reactions were poorly documented. With increased interests in cellulases in bio-fuel industry, the prediction of parameters in enzymatic reactions is listed on agenda. In this study, two aims were defined (a) which amino acid property works better to predict the temperature optimum and (b) with which deletion the jackknife validation works. The results showed that the amino acid distribution probability works better in predicting the optimum temperature of catalytic reaction by cellulase, and the delete-4, more precisely one-fifth deletion, jackknife validation works better.

Universal peptide vaccines - optimal peptide vaccine design based on viral sequence conservation

Rapidly mutating viruses such as the hepatitis C virus (HCV), the human immunodeficiency virus (HIV), or influenza viruses (Flu) call for highly effective universal peptide vaccines, i.e. vaccines that do not only yield broad population coverage but also broad coverage of various viral strains. The efficacy of such vaccines is determined by multiple properties of the epitopes they comprise. Beyond the specific properties of each epitope, properties of the corresponding source antigens are of great importance. If a response is mounted against viral proteins with a low copy number within the cell or against proteins expressed very late, this response may fail to induce lysis of the infected cells before budding can take place. We here propose a novel methodology to optimize the epitope composition and assembly in order to induce maximum protection. In order for a peptide vaccine to yield the best possible universal protection, several conditions should be met: (a) an optimal choice of target antigens, (b) an optimal choice of highly conserved epitopes, (c) maximum coverage of the target population, and (d) the proper ordering of the epitopes in the final vaccine to ensure favorable cleavage. We propose a mathematical formalism for epitope selection and ordering that balances the constraints imposed by these different conditions. Focusing on HCV, HIV, and Flu, we show that not all of the conditions can be satisfied for all viruses. Depending on the virus, different constraints are harder to fulfill: for Flu, the conservation constraint is violated first, while for HIV, it is difficult to focus the response at the optimal target antigens. The proposed methodology can be applied to any virus to assess the feasibility of optimally combining the above-mentioned constraints.

Fitting evolutionary process of influenza A virus nucleoproteins using analytical solution of system of differential equations

Very recently we explored the possibility of using differential equations to describe the evolution of proteins. In this study we used the amino-acid pair predictability to quantify 1709 nucleoproteins of influenza A viruses isolated from 1918 to 2008 to represent their evolutionary process, thereafter we used the analytical solution of system of differential equations to fit the evolution of the nucleoprotein family. The results showed that the analytical solution could fit the nucleoprotein evolution and the obtained parameters were useful for timing of future mutations. Our approach provided a way to quantitatively analyze protein dynamics and evolution.

Evidence obtained from ANOVA to reason cross-species infection and cross-subtype mutation in neuraminidases of influenza A viruses

The current pandemic of A/H1N1 influenza raises a serious question on cross-species infection and cross-subtype mutation because our previous focus on possible influenza pandemic laid on H5N1 subtype and the cross-species infection between avian and human. In this study, we analyse 3874 neuraminidases from influenza A viruses using anova to answer the question of if there is barrier between species and between subtypes. The results show that there is no cross-species barrier in some species, and the intra-species variation is larger than the inter-species variation in some species hosting the viruses, and the cross-subtype mutation is possible because there is no cross-subtype barrier in some subtypes and the intra-subtype variation is larger than the inter-subtype variation in some subtypes. These results highlight the state of barrier of influenza A virus, which can help us understand the current pandemic and manufacture more effective vaccines.

Mutation patterns in human α-galactosidase A

A way to study the mutation pattern is to convert a 20-letter protein sequence into a scalar protein sequence, because the 20-letter protein sequence is neither vector nor scalar while a promising way to study patterns is in numerical domain. In this study, we use the amino-acid pair predictability to convert α-galactosidase A with its 137 mutations into scalar sequences, and analyse which amino-acid pairs are more sensitive to mutation. Our results show that the unpredictable amino-acid pairs are more sensitive to mutation, and the mutation trend is to narrow the difference between predicted and actual frequency of amino-acid pairs.

Trends in global warming and evolution of matrix protein 2 family from influenza A virus

The global warming is an important factor affecting the biological evolution, and the influenza is an important disease that threatens humans with possible epidemics or pandemics. In this study, we attempted to analyze the trends in global warming and evolution of matrix protein 2 family from influenza A virus, because this protein is a target of anti-flu drug, and its mutation would have significant effect on the resistance to anti-flu drugs. The evolution of matrix protein 2 of influenza A virus from 1959 to 2008 was defined using the unpredictable portion of amino-acid pair predictability. Then the trend in this evolution was compared with the trend in the global temperature, the temperature in north and south hemispheres, and the temperature in influenza A virus sampling site, and species carrying influenza A virus. The results showed the similar trends in global warming and in evolution of M2 proteins although we could not correlate them at this stage of study. The study suggested the potential impact of global warming on the evolution of proteins from influenza A virus.

Quantitative relationship between mutated amino-acid sequence of human copper-transporting ATPases and their related diseases

Copper-transporting ATPase 1 and 2 (ATP7A and ATP7B) are two highly homologous P-type copper ATPase exporters. Mutations in ATP7A can lead to Menkes disease which is an X-linked disorder of copper deficiency. Mutations in ATP7B can cause Wilson disease which is an autosomal recessive disorder of copper toxicity. In this study, we attempt to build a quantitative relationship between mutated ATPase and Menkes/Wilson disease. First, we use the amino-acid distribution probability as a measure to quantify the difference in ATPase before and after mutation. Second, we use the cross-impact analysis to define the quantitative relationship between mutant ATPase protein and Menkes/Wilson disease, and compute various probabilities. Finally, we use the Bayesian equation to determine the probability that Menkes/Wilson disease is diagnosed under a mutation. The results show (i) the vast majority of mutations lead to the amino-acid distribution probability increase in mutant ATP7As and decrease in ATP7Bs, and (ii) the probability that a mutation causes Menkes/Wilson disease is about nine tenth. Thus we provide a way to use the descriptively probabilistic method to couple the mutation with its clinical outcome after quantifying mutations in proteins.

The Influenza Primer Design Resource: a new tool for translating influenza sequence data into effective diagnostics

BACKGROUND

Recent outbreaks of highly pathogenic avian influenza and multiple occurrences of zoonotic infection and deaths in humans have sparked a dramatic increase in influenza research. In order to rapidly identify and help prevent future influenza outbreaks, numerous laboratories around the world are working to develop new nucleotide-based diagnostics for identifying and subtyping influenza viruses. While there are several databases that have been developed for manipulating the vast amount of influenza genetic data that have been produced, significant progress can still be made in developing tools for translating the genetic data into effective diagnostics.

DESCRIPTION

The Influenza Primer Design Resource (IPDR) is the combination of a comprehensive database of influenza nucleotide sequences and a web interface that provides several important tools that aid in the development of oligonucleotides that may be used to develop better diagnostics. IPDR's database can be searched using a variety of criteria, allowing the user to align the subset of influenza sequences that they are interested in. In addition, IPDR reports a consensus sequence for the alignment along with sequence polymorphism information, a summary of most published primers and probes that match the consensus sequence, and a Primer3 analysis of potential primers and probes that could be used for amplifying the sequence subset.

CONCLUSIONS

The IPDR is a unique combination of bioinformatics tools that will greatly aid researchers in translating influenza genetic data into diagnostics, which can effectively identify and subtype influenza strains. The website is freely available at http://www.ipdr.mcw.edu.

Prediction of mutations engineered by randomness in H5N1 hemagglutinins of influenza A virus

This is the continuation of our studies on the prediction of mutation engineered by randomness in proteins from influenza A virus. In our previous studies, we have demonstrated that randomness plays a role in engineering mutations because the measures of randomness in protein are different before and after mutations. Thus we built a cause-mutation relationship to count the mutation engineered by randomness, and conducted several concept-initiated studies to predict the mutations in proteins from influenza A virus, which demonstrated the possibility of prediction of mutations along this line of thought. On the other hand, these concept-initiated studies indicate the directions forwards the enhancement of predictability, of which we need to use the neural network instead of logistic regression that was used in those concept-initiated studies to enhance the predictability. In this proof-of-concept study, we attempt to apply the neural network to modeling the cause-mutation relationship to predict the possible mutation positions, and then we use the amino acid mutating probability to predict the would-be-mutated amino acids at predicted positions. The results confirm the possibility of use of internal cause-mutation relationship with neural network model to predict the mutation positions and use of amino acid mutating probability to predict the would-be-mutated amino acids.

Prediction of mutations engineered by randomness in H5N1 neuraminidases from influenza A virus

In this proof-of-concept study, we attempt to determine whether the cause-mutation relationship defined by randomness is protein dependent by predicting mutations in H5N1 neuraminidases from influenza A virus, because we have recently conducted several concept-initiated studies on the prediction of mutations in hemagglutinins from influenza A virus. In our concept-initiated studies, we defined the randomness as a cause for mutation, upon which we built a cause-mutation relationship, which is then switched into the classification problem because the occurrence and non-occurrence of mutations can be classified as unity and zero. Thereafter, we used the logistic regression and neural network to solve this classification problem to predict the mutation positions in hemagglutinins, and then used the amino acid mutating probability to predict the would-be-mutated amino acids. As the previous results were promising, we extend this approach to other proteins, such as H5N1 neuraminidase in this study, and further address various issues raised during the development of this approach. The result of this study confirms that we can use this cause-mutation relationship to predict the mutations in H5N1 neuraminidases.