A comparative analysis of computational tools for the prediction of epigenetic DNA methylation from long-read sequencing data
Pai, S. S.; Mathew, A. R.; Anindya, R.
Show abstract
Recent development of Oxford Nanopore long-read sequencing has opened new avenues of identifying epigenetic DNA methylation. Among the different epigenetic DNA methylations, N6-methyladenosine is the most prevalent DNA modification in prokaryotes and 5-methylcytosine is common in higher eukaryotes. Here we investigated if N6-methyladenosine and 5-methylcytosine modifications could be predicted from the nanopore sequencing data. Using publicly available genome sequencing data of Saccharomyces cerevisiae, we compared the open-access computational tools, including Tombo, mCaller, Nanopolish and DeepSignal for predicting 6mA and 5mC. Our results suggest that Tombo and mCaller can predict DNA N6-methyladenosine modifications at a specific location, whereas, Tombo dampened fraction, Nanopolish methylation likelihood and DeepSignal methylation probability have comparable efficiency for 5-methylcytosine prediction from Oxford Nanopore sequencing data.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- MethPanel: a parallel pipeline and interactive analysis tool for multiplex bisulphite PCR sequencing to assess DNA methylation biomarker panels for disease detection 95%
- Snapper: a high-sensitive algorithm to detect methylation motifs based on Oxford Nanopore reads 94%
- Benchmarking Peak Calling Methods for CUT&RUN 93%
Similar papers in this journal
Similar papers in this journal
- DeepPlnc: Bi-modal Deep Learning for Highly Accurate Plant lncRNA Discovery 93%
- Principal component analysis- and tensor decomposition-based unsupervised feature extraction to select more reasonable differentially methylated cytosines: Optimization of standard deviation versus state-of-the-art methods 93%
- Minimum Error Calibration and Normalization for Genomic Copy Number Analysis 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.