Abstract
For metagenomics datasets, datasets of complex polyploid genomes, and other high-variation genomics datasets, there are difficulties with the analysis, error detection and variant calling, stemming from the challenges of discerning sequencing errors from biological variation. Confirming base candidates with high frequency of occurrence is no longer a reliable measure because of the natural variation and the presence of rare bases. The paper discusses an approach to the application of machine learning models to classify bases into erroneous and rare variations after preselecting potential error candidates with a weighted frequency measure, which aims to focus on unexpected variations by using the inter-sequence pairwise similarity. Different similarity measures are used to account for different types of datasets. Four machine learning models are implemented and tested.
Highlights
A lot of genomics research has to deal with high-variation datasets
This paper presents a combination of two approaches to identify errors in high-variation datasets, such as found in metagenomics and polyploid genome analysis
The first analytical approach relies on the introduction of weighted frequencies, while the second trains different machine learning models to perform the error reclassification
Summary
A lot of genomics research has to deal with high-variation datasets This includes metagenomics [1], where it is common to work with the genomes of multiple organisms in the same sample, and polyploid species [2] that have more than one similar and hard-to-differentiate subgenomes. Metagenomics studies microorganism communities in environmental samples These range from soil samples in agriculture [3] to human microbiome in health research [4]. Whether these studies have a specific focus on the individual variations, or a general focus on the diversity and interspecies evolutionary relations, they all face difficulties because of the errors in the datasets [5], which has led to a common practice of discarding entire sequences suspected to contain errors [6]. For error correction to be applicable in those fields, errors in the experimental reading need to be correctly distinguished from the natural sources of variation, such as single-point polymorphisms (SNPs), which account for genuine differences between biological organisms, and are the target of research whose preservation is necessary
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
