DNAscent

DNAscent detects BrdU base analogues in Oxford Nanopore Technologies (ONT) sequencing reads to map DNA replication dynamics at single-molecule, single-base resolution.


Key Features:

  • Base analogue detection: Detects the thymidine analogue BrdU within ONT sequencing reads.
  • Residual neural network: Uses a residual neural network for fast, high-precision identification of BrdU at single-base resolution.
  • Autoencoder-based replication feature detection: Employs an autoencoder to detect replication forks, origins, and termination sites.
  • Single-molecule and single-base resolution: Reports replication-related signals at the level of individual DNA molecules and individual bases.
  • High-throughput analysis: Processes large ONT datasets to enable measurement of replication features across many reads.
  • Compatibility with experimental designs: Applicable to both synchronous and asynchronous cell population experiments.
  • Protocol robustness: Demonstrates robustness across various experimental protocols for BrdU incorporation and ONT sequencing.

Scientific Applications:

  • DNA replication dynamics: Mapping replication fork movement and timing to study replication dynamics genome-wide.
  • Origin and termination site mapping: Identifying replication origins and termination sites from single-molecule ONT data.
  • Base analogue incorporation analysis: Quantifying and localizing BrdU incorporation patterns in synchronous and asynchronous cell populations.

Methodology:

DNAscent applies deep learning to ONT sequencing reads, using a residual neural network to detect BrdU and an autoencoder to identify replication forks, origins, and termination sites.

Topics

Details

License:
GPL-3.0
Programming Languages:
C++, Python
Added:
1/18/2021
Last Updated:
3/1/2021

Operations

Publications

Boemo MA. DNAscent v2: Detecting Replication Forks in Nanopore Sequencing Data with Deep Learning. Unknown Journal. 2020. doi:10.1101/2020.11.04.368225.

Documentation