Benchmarking machine learning robustness in Covid-19 genome sequence classification.

The rapid spread of the COVID-19 pandemic has resulted in an unprecedented amount of sequence data of the SARS-CoV-2 genome-millions of sequences and counting. This amount of data, while being orders of magnitude beyond the capacity of traditional approaches to understanding the diversity, dynamics, and evolution of viruses, is nonetheless a rich resource for machine learning (ML) approaches as alternatives for extracting such important information from these data. It is of hence utmost importance to design a framework for testing and benchmarking the robustness of these ML models. This paper makes the first effort (to our knowledge) to benchmark the robustness of ML models by simulating biological sequences with errors. In this paper, we introduce several ways to perturb SARS-CoV-2 genome sequences to mimic the error profiles of common sequencing platforms such as Illumina and PacBio. We show from experiments on a wide array of ML models that some simulation-based approaches with different perturbation budgets are more robust (and accurate) than others for specific embedding methods to certain noise simulations on the input sequences. Our benchmarking framework may assist researchers in properly assessing different ML models and help them understand the behavior of the SARS-CoV-2 virus or avoid possible future pandemics.

Efficient Approximate Kernel Based Spike Sequence Classification.

Machine learning (ML) models, such as SVM, for tasks like classification and clustering of sequences, require a definition of distance/similarity between pairs of sequences. Several methods have been proposed to compute the similarity between sequences, such as the exact approach that counts the number of matches between k-mers (sub-sequences of length k) and an approximate approach that estimates pairwise similarity scores. Although exact methods yield better classification performance, they pose high computational costs, limiting their applicability to a small number of sequences. The approximate algorithms are proven to be more scalable and perform comparably to (sometimes better than) the exact methods - they are designed in a "general" way to deal with different types of sequences (e.g., music, protein, etc.). Although general applicability is a desired property of an algorithm, it is not the case in all scenarios. For example, in the current COVID-19 (coronavirus) pandemic, there is a need for an approach that can deal specifically with the coronavirus. To this end, we propose a series of ways to improve the performance of the approximate kernel (using minimizers and information gain) in order to enhance its predictive performance pm coronavirus sequences. More specifically, we improve the quality of the approximate kernel using domain knowledge (computed using information gain) and efficient preprocessing (using minimizers computation) to classify coronavirus spike protein sequences corresponding to different variants (e.g., Alpha, Beta, Gamma). We report results using different classification and clustering algorithms and evaluate their performance using multiple evaluation metrics. Using two datasets, we show that our proposed method helps improve the kernel's performance compared to the baseline and state-of-the-art approaches in the healthcare domain.

From Alpha to Zeta: Identifying Variants and Subtypes of SARS-CoV-2 Via Clustering.

The availability of millions of SARS-CoV-2 (Severe Acute Respiratory Syndrome-Coronavirus-2) sequences in public databases such as GISAID (Global Initiative on Sharing All Influenza Data) and EMBL-EBI (European Molecular Biology Laboratory-European Bioinformatics Institute) (the United Kingdom) allows a detailed study of the evolution, genomic diversity, and dynamics of a virus such as never before. Here, we identify novel variants and subtypes of SARS-CoV-2 by clustering sequences in adapting methods originally designed for haplotyping intrahost viral populations. We asses our results using clustering entropy-the first time it has been used in this context. Our clustering approach reaches lower entropies compared with other methods, and we are able to boost this even further through gap filling and Monte Carlo-based entropy minimization. Moreover, our method clearly identifies the well-known Alpha variant in the U.K. and GISAID data sets, and is also able to detect the much less represented (<1% of the sequences) Beta (South Africa), Epsilon (California), and Gamma and Zeta (Brazil) variants in the GISAID data set. Finally, we show that each variant identified has high selective fitness, based on the growth rate of its cluster over time. This demonstrates that our clustering approach is a viable alternative for detecting even rare subtypes in very large data sets.