Convolutional Classification of Pathogenicity in H5 Avian Influenza Strains

Chadha, Akshay; Dara, Rozita; Poljak, Zvonimir

doi:10.1109/icmla.2019.00259

Cited by 7 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among them are protein structure prediction [39], gene expression regulation [40][41][42], predicting the sequence specificities [43], and protein classification [44]. Recently, deep learning has been applied to predict the mutation of the influenza virus [45], pathogenicity classification of H5 avian influenza [46], as well as time-series modeling for the recently emerging COVID-19 outbreak [47]. An inevitable problem in omics research is the representation of raw biological sequences, that is, amino acid sequence, as a network input.…”

Section: Related Workmentioning

confidence: 99%

Convolutional Neural Network Based Approach to In Silico Non-Anticipating Prediction of Antigenic Distance for Influenza Virus

Forghani

Khachay

2020

Viruses

View full text Add to dashboard Cite

Evaluation of the antigenic similarity degree between the strains of the influenza virus is highly important for vaccine production. The conventional method used to measure such a degree is related to performing the immunological assays of hemagglutinin inhibition. Namely, the antigenic distance between two strains is calculated on the basis of HI assays. Usually, such distances are visualized by using some kind of antigenic cartography method. The known drawback of the HI assay is that it is rather time-consuming and expensive. In this paper, we propose a novel approach for antigenic distance approximation based on deep learning in the feature spaces induced by hemagglutinin protein sequences and Convolutional Neural Networks (CNNs). To apply a CNN to compare the protein sequences, we utilize the encoding based on the physical and chemical characteristics of amino acids. By varying (hyper)parameters of the CNN architecture design, we find the most robust network. Further, we provide insight into the relationship between approximated antigenic distance and antigenicity by evaluating the network on the HI assay database for the H1N1 subtype. The results indicate that the best-trained network gives a high-precision approximation for the ground-truth antigenic distances, and can be used as a good exploratory tool in practical tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

Convolutional Neural Network Based Approach to In Silico Non-Anticipating Prediction of Antigenic Distance for Influenza Virus

Forghani

Khachay

2020

Viruses

View full text Add to dashboard Cite

show abstract

“…The rationale behind this decision was to sample some of the genes that we had worked on in our previous research endeavors [15], Our objective was to integrate automated learning algorithms and pattern-matching algorithms that are based on specific DNA sequences, in order to create a biological data collection that could be utilized in a classification process. We conducted experiments on a dataset that included DNA sequences, where we compared the effectiveness of searching for a specific pattern with other classification models, such as Random Forest [3,16], KNN [16][17][18][19][20], Naïve Bayes [21][22][23][24], Decision tree [23,[25][26][27][28][29][30], and Support Vector Machine [18,[31][32][33][34][35][36] with Linear [37,38], RBF [37,39], and sigmoid [21,40] classifiers, the results of these classifiers models are calculated by F1 score, recall, precision rate, execution time, and with the accuracy which calculates the most effective patternmatching classifier. The comparison of DNA sequences is a crucial task in various fields of research, including molecular biology and genetics.…”

Section: Methodology For Pm From Dna Sequencesmentioning

confidence: 99%

Optimizing classification efficiency with machine learning techniques for pattern matching

Hamed¹,

Ibrahim²,

El‐Hafeez³

2023

J Big Data

View full text Add to dashboard Cite

The study proposes a novel model for DNA sequence classification that combines machine learning methods and a pattern-matching algorithm. This model aims to effectively categorize DNA sequences based on their features and enhance the accuracy and efficiency of DNA sequence classification. The performance of the proposed model is evaluated using various machine learning algorithms, and the results indicate that the SVM linear classifier achieves the highest accuracy and F1 score among the tested algorithms. This finding suggests that the proposed model can provide better overall performance than other algorithms in DNA sequence classification. In addition, the proposed model is compared to two suggested algorithms, namely FLPM and PAPM, and the results show that the proposed model outperforms these algorithms in terms of accuracy and efficiency. The study further explores the impact of pattern length on the accuracy and time complexity of each algorithm. The results show that as the pattern length increases, the execution time of each algorithm varies. For a pattern length of 5, SVM Linear and EFLPM have the lowest execution time of 0.0035 s. However, at a pattern length of 25, SVM Linear has the lowest execution time of 0.0012 s. The experimental results of the proposed model show that SVM Linear has the highest accuracy and F1 score among the tested algorithms. SVM Linear achieved an accuracy of 0.963 and an F1 score of 0.97, indicating that it can provide the best overall performance in DNA sequence classification. Naive Bayes also performs well with an accuracy of 0.838 and an F1 score of 0.94. The proposed model offers a valuable contribution to the field of DNA sequence analysis by providing a novel approach to pre-processing and feature extraction. The model’s potential applications include drug discovery, personalized medicine, and disease diagnosis. The study’s findings highlight the importance of considering the impact of pattern length on the accuracy and time complexity of DNA sequence classification algorithms.

show abstract

“…The model produced predictions with a test accuracy score of 0.966. Chadha et al (2019) developed a model to classify H5 avian IAVs based on high and low pathogenicity [44]. Via a convolutional NN, the test accuracy achieved was 0.992. compared amino acid positions identified by RF and ADABOOST to a list of known markers of infectivity, transmissibility, and pathogenicity [28].…”

Section: Characteristics Of Infectionmentioning

confidence: 99%

Influenza virus genotype to phenotype predictions through machine learning: a systematic review

Borkenhagen

Allen

Runstadler

2021

Emerging Microbes & Infections

View full text Add to dashboard Cite

Background: There is great interest in understanding the viral genomic predictors of phenotypic traits that allow influenza A viruses to adapt to or become more virulent in different hosts.Machine learning techniques have demonstrated promise in addressing this critical need for other pathogens because the underlying algorithms are especially well equipped to uncover complex patterns in large datasets and produce generalizable predictions for new data. As the body of research where these techniques are applied for influenza A virus phenotype prediction continues to grow, it is useful to consider the strengths and weaknesses of these approaches to understand what has prevented these models from seeing widespread use by surveillance laboratories and to identify gaps that are underexplored with this technology. Methods and Results: We present a systematic review of English literature published to date of studies employing machine learning methods to generate predictions of influenza A virus phenotypes from genomic or proteomic input. Forty nine studies were included in this review, spanning the topics of host discrimination, human adaptability, subtype and clade assignment, pandemic lineage assignment, characteristics of infection, and antiviral drug resistance.Conclusions: Our findings suggest that biases in model design and a dearth of wet laboratory follow-up may explain why these models often go underused. We, therefore, offer guidance to overcome these limitations, aid in improving predictive models of previously studied influenza A virus phenotypes, and extend those models to unexplored phenotypes in the ultimate pursuit of tools to enable the characterization of virus isolates across surveillance laboratories.

show abstract

Convolutional Classification of Pathogenicity in H5 Avian Influenza Strains

Cited by 7 publications

References 26 publications

Convolutional Neural Network Based Approach to In Silico Non-Anticipating Prediction of Antigenic Distance for Influenza Virus

Convolutional Neural Network Based Approach to In Silico Non-Anticipating Prediction of Antigenic Distance for Influenza Virus

Optimizing classification efficiency with machine learning techniques for pattern matching

Influenza virus genotype to phenotype predictions through machine learning: a systematic review

Contact Info

Product

Resources

About