Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

Matthew T. Parker; Geoffrey J. Barton; Gordon G. Simpson

doi:10.1101/2020.05.27.118679

Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

Matthew T. Parker (Lead / Corresponding author)
, Geoffrey J. Barton
, Gordon G. Simpson (Lead / Corresponding author)

Research output: Working paper/Preprint › Preprint

Abstract

Transcription of eukaryotic genomes involves complex alternative processing of RNAs. Sequencing of full-length RNAs using long-reads reveals the true complexity of processing, however the relatively high error rates of long-read technologies can reduce the accuracy of intron identification. Here we present a two-pass approach, combining alignment metrics and machine-learning-derived sequence information to filter spurious examples from splice junctions identified in long-read alignments. The remaining junctions are then used to guide realignment. This method, available in the software package 2passtools (https://github.com/bartongroup/2passtools), improves the accuracy of spliced alignment and transcriptome annotation without requiring orthogonal information from short read RNAseq or existing annotations.

Original language	English
Place of Publication	Cold Spring Harbor Laboratory
Publisher	BioRxiv
Number of pages	36
DOIs	https://doi.org/10.1101/2020.05.27.118679
Publication status	Published - 30 May 2020

Access to Document

10.1101/2020.05.27.118679Licence: CC BY-ND

2passtools: two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing
Parker, M. T. (Lead / Corresponding author), Knop, K., Barton, G. J. & Simpson, G. G. (Lead / Corresponding author), 1 Mar 2021, In: Genome Biology. 22, 1, 24 p., 72.
Research output: Contribution to journal › Article › peer-review

Open Access
File
17 Link opens in a new tab Citations (Scopus)

310 Downloads (Pure)

Cite this

@techreport{aae011cf05b0478a89470bceb26a82fa,

title = "Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing",

abstract = "Transcription of eukaryotic genomes involves complex alternative processing of RNAs. Sequencing of full-length RNAs using long-reads reveals the true complexity of processing, however the relatively high error rates of long-read technologies can reduce the accuracy of intron identification. Here we present a two-pass approach, combining alignment metrics and machine-learning-derived sequence information to filter spurious examples from splice junctions identified in long-read alignments. The remaining junctions are then used to guide realignment. This method, available in the software package 2passtools (https://github.com/bartongroup/2passtools), improves the accuracy of spliced alignment and transcriptome annotation without requiring orthogonal information from short read RNAseq or existing annotations.",

author = "Parker, \{Matthew T.\} and Barton, \{Geoffrey J.\} and Simpson, \{Gordon G.\}",

year = "2020",

month = may,

day = "30",

doi = "10.1101/2020.05.27.118679",

language = "English",

publisher = "BioRxiv",

address = "United States",

type = "WorkingPaper",

institution = "BioRxiv",

}

TY - UNPB

T1 - Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

AU - Parker, Matthew T.

AU - Barton, Geoffrey J.

AU - Simpson, Gordon G.

PY - 2020/5/30

Y1 - 2020/5/30

N2 - Transcription of eukaryotic genomes involves complex alternative processing of RNAs. Sequencing of full-length RNAs using long-reads reveals the true complexity of processing, however the relatively high error rates of long-read technologies can reduce the accuracy of intron identification. Here we present a two-pass approach, combining alignment metrics and machine-learning-derived sequence information to filter spurious examples from splice junctions identified in long-read alignments. The remaining junctions are then used to guide realignment. This method, available in the software package 2passtools (https://github.com/bartongroup/2passtools), improves the accuracy of spliced alignment and transcriptome annotation without requiring orthogonal information from short read RNAseq or existing annotations.

AB - Transcription of eukaryotic genomes involves complex alternative processing of RNAs. Sequencing of full-length RNAs using long-reads reveals the true complexity of processing, however the relatively high error rates of long-read technologies can reduce the accuracy of intron identification. Here we present a two-pass approach, combining alignment metrics and machine-learning-derived sequence information to filter spurious examples from splice junctions identified in long-read alignments. The remaining junctions are then used to guide realignment. This method, available in the software package 2passtools (https://github.com/bartongroup/2passtools), improves the accuracy of spliced alignment and transcriptome annotation without requiring orthogonal information from short read RNAseq or existing annotations.

U2 - 10.1101/2020.05.27.118679

DO - 10.1101/2020.05.27.118679

M3 - Preprint

BT - Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

PB - BioRxiv

CY - Cold Spring Harbor Laboratory

ER -

Two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

Abstract

Access to Document

Fingerprint

Research output

2passtools: two-pass alignment using machine-learning-filtered splice junctions increases the accuracy of intron detection in long-read RNA sequencing

Cite this