seqfold 0.3.2

Predict the minimum free energy structure of nucleic acids

seqfold

Predict the minimum free energy structure of nucleic acids.

seqfold is an implementation of the Zuker, 1981 dynamic programming algorithm, the basis for UNAFold/mfold, plus energy functions from SantaLucia, 2004.

Installation

pip install seqfold

Usage

Python

from seqfold import calc_dg

# a bifurcated DNA structure
calc_dg("GGGAGGTCGTTACATCTGGGTAACACCGGTACTGATCCGGTGACCTCCC")  # -12.94

CLI

$ seqfold TAGCTCAGCTGGGAGAGCGCCTGCTTTGCACGCAGGAGGT -t 32
-6.58

Motivation

Knowing nucleic acid sequences' secondary structures is essential in synbio for selecting primers for PCR, designing oligos for MAGE, and tuning RBS expression rates.

While UNAFold and mfold are the most widely used applications for nucleic acid secondary structure prediction, their format and license are restrictive. seqfold is meant to be a more open-source, but minimal, application for predicting minimum free energy secondary structure.

	seqfold	mfold	UNAFold
License	MIT	Academic Non-commercial	$200-36,000
OS	Linux, MacOS, Windows	Linux, MacOS	Linux, MacOS, Windows
Format	python, CLI python	CLI binary	CLI binary
Dependencies	none	(mfold_util)	Perl, (gnuplot, glut/OpenGL)
Graphical	no	yes (output)	yes (output)
Heterodimers	no	yes	yes
Constraints	no	yes	yes

Citations

That papers and others that were used to develop this library are below. Each paper is listed along with how it relates to seqfold.

Nussinov, 1980

Nussinov, Ruth, and Ann B. Jacobson. "Fast algorithm for predicting the secondary structure of single-stranded RNA." Proceedings of the National Academy of Sciences 77.11 (1980): 6309-6313.

Framework for the dynamic programming approach. It has a conceptually helpful "Maximal Matching" example that demonstrates the approach on a simple sequence with only matched or unmatched bp.

Zuker, 1981

Zuker, Michael, and Patrick Stiegler. "Optimal computer folding of large RNA sequences using thermodynamics and auxiliary information." Nucleic acids research 9.1 (1981): 133-148.

The most cited paper in this space. Extends further than Nussinov, 1980 with a nearest neighbor approach to energies and a consideration of each of stack, bulge, internal loop, and hairpin. Their data structure and traceback method are both more intuitive than Nussinov, 1980.

Jaeger, 1989

Jaeger, John A., Douglas H. Turner, and Michael Zuker. "Improved predictions of secondary structures for RNA." Proceedings of the National Academy of Sciences 86.20 (1989): 7706-7710.

Zuker and colleagues expand on the 1981 paper to incorporate penalties for multibranched loops and dangling ends.

SantaLucia, 2004

SantaLucia Jr, John, and Donald Hicks. "The thermodynamics of DNA structural motifs." Annu. Rev. Biophys. Biomol. Struct. 33 (2004): 415-440.

The paper from which almost every DNA energy function in seqfold comes from (with the exception of multibranch loops). Provides neighbor entropies and enthalpies for stacks, mismatching stacks, terminal stacks, and dangling stacks. Ditto for bulges, internal loops, and hairpins.

Turner, 2009

Turner, Douglas H., and David H. Mathews. "NNDB: the nearest neighbor parameter database for predicting stability of nucleic acid secondary structure." Nucleic acids research 38.suppl_1 (2009): D280-D282.

Source of RNA nearest neighbor change in entropy and enthalpy parameter data. In /data.

Ward, 2017

Ward, M., Datta, A., Wise, M., & Mathews, D. H. (2017). Advanced multi-loop algorithms for RNA secondary structure prediction reveal that the simplest model is best. Nucleic acids research, 45(14), 8541-8550.

An investigation of energy functions for multibranch loops that validates the simple linear approach employed by Jaeger, 1989 that keeps runtime at O(n³).