RNA secondary structure prediction engine

The monorepo provides a grammar based algorithm to predict the pseudoknots patters of the secondary structure of any RNA sequence.

Authors: Andrikos Christos, Makris Evaggelos, Pavlatos Christos, Rassias Georgios, Aggelos Kolaitis

Implementation details

The algorighm consits of the two following steps:

1.  Parse mulitple RNA subsequences to define the potenital pseudoknot structures

2.  Choose the pseudoknot structure that is considered to be the most stable one. This steps is based on the concept of energy minimization concept.

The core algorithm was initially implemented in python based on the wide-known NLTK package. Due to serious performance issues we moved the parsing into c utilizing the yaep parser which is able to parse ambient grammars. The operation goes as it follows:

1.
2.
2.
2.
2.
2.

Scoring

Compare prediction dot bracket with ground truth. Create confusion matrix

Definition	Description
true positive	Ground truth has a stem here, and I have correctly found that stem (matching the pair, TODO: distance +- 1)
true negative	Ground truth does not have a stem, and I do not have a stem
false positive	I have predicted a stem here, but there is no stem in ground truth
false negative	I have predicted no stem, but ground truth has a stem here

Development Environment

Building the code consists of 2 parts: Setting up a Python 3 virtual environment and building the C parser library.

$ make deps  # install package dependencies
$ make       # build the parser and setup the virtual environment at ./.venv

Run for all cases and save result to result.yaml:

$ ./.venv/bin/rna_benchmark --cases cases.yaml --grammar ./libpseudoknot.so --max-dd-size 2 --max-stem-allow-smaller 1 --allow-ug --prune-early > result.yaml

Execute

For a single sequence. See --help for a complete list of options:

$ ./.venv/bin/rna_analysis --grammar ./libpseudoknot.so AAAAAACUAAUAGAGGGGGGACUUAGCGCCCCCCAAACCGUAACCCC

Run benchmark for a number of cases, print output in YAML file. See cases.yaml for an example YAML file:

$ ./.venv/bin/rna_benchmark --grammar ./libpseudoknot.so [OPTIONS] --cases cases.yaml > results.yaml

Unit Tests

Activate virtual environment and run with Pytest:

$ ./.venv/bin/pytest

Loop sizes and indices

Unused right loop:

• inclusive_start_index = left_core_inner + left_loop_stems + 1
• unused_right_loop_size = right_core_outer - inclusive_start_index
• inclusive_end_index = right_core_outer - 1

Unused left loop:

• unused_left_loop_size = left_loop_size - right_core_stems
• inclusive_start_index = left_core_outer + 1
• inclusive_end_index = inclusive_start_index + left_loop_size - right_core_stems - 1