Skip to content

SangTK is an ML-backed toolkit that improves the quality of Sanger base calls.

Notifications You must be signed in to change notification settings

asan-emirsaleh/SangTK

Folders and files

NameName
Last commit message
Last commit date

Latest commit

9201423 · Mar 10, 2020

History

29 Commits
Nov 4, 2019
Nov 3, 2019
Nov 4, 2019
Nov 3, 2019
Nov 4, 2019
Feb 21, 2020
Oct 30, 2019
Nov 4, 2019
Aug 23, 2019
Oct 17, 2019
Oct 6, 2019
Nov 3, 2019

Repository files navigation

SangTK

SangTK is an ML-backed toolkit that improves the quality of Sanger base calls.

Function

SangTK is a toolkit for AB1 (sanger sequencing) files that can perform any combination of the following functions:

  • Convert .abi files to .fa files
  • Convert a directory of .abi files into a single fa file or separate .fa files
  • Perform logistic regression on nucleotide peaks to improve sequence quality
  • Call peaks using a Bi-Directional LSTM RNN
  • Improve the quality of the supplied peak calls

Dependencies

  • python 3
  • biopython
  • pandas
  • sklearn
  • tensorflow/keras
  • scipy
  • numpy
  • ncbi blastn (only for testing)

Usage

python sang.py [-h]

Essential (exclusive) inputs

-d/--ab1_directory
Directory containing .ab1 files to be converted into fasta file.

-f/--ab1_file
Single .ab1 file to be converted into fasta file.

-t/--testing
Testing functionality. Use either 'nucleotide' to test nucleotide calling or 'peak' to test both nucleotide and peak testing.

Optional inputs

-s/--split
Split output into separate fasta files. Input as true or false.

-o/--fa_name
Input non-default name for fasta file (not valid if inputting >1 .ab1 file with -s flag).

-pn/--predict_nucleotide
Use predictive algorithm to determine sequence. Calls nucleotides given peaks in .ab1 file.

-p/--predict_peak_and_nucleotide
Use predictive algorithm to determine sequence. Calls both peaks and nucleotides.

About

SangTK is an ML-backed toolkit that improves the quality of Sanger base calls.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published