Author: Fang-Chieh Chou

May 2015 by Fang-Chieh Chou (fcchou [at] stanford.edu).

# Code and Demo

The main RECCES application is main/source/src/apps/public/rna_util/recces_turner.cc. It is accompanied by a set of python codes in tools/recces. A README file for the python codes is included. The optimized RECCES score function is main/database/scoring/stepwise/rna/turner.wts.

For a minimal demonstration of RECCES, see: demos/public/recces/. Online documentation for the RECCES demo is also available.

# Application purpose

This code provides a way to compute the free energy of an RNA molecule using comprehensive sampling to account for the conformational entropy. RECCES also allows rapid reweighting of the score function by caching the sub-scores of each sampled conformation.

# Algorithm

RECCES uses simulated-tempering Monte Carlo methods to efficiently sample the conformational ensemble. Standard Rosetta score terms are used for the calculation; the terms are then reweighted to fit against experimental folding

# Limitations

• RECCES currently works for RNA duplexes and dangling-ends only. While it is possible to extend the framework to other non-canonical RNA motifs and even protein applications, such work has not yet been performed.

• The score terms being cached are currently hard-coded in the source code (recces_turner.cc) and the Python scripts; therefore adding new score terms requires editing the codes, which is not convenient. This can be make more general in the future by including a current_score_terms file for both the Rosetta and Python codes.

# Modes

There is only one mode to run RECCES at present.

# Input Files

There is no specific input file required RECCES. One may use a different score function file for the simulated tempering simulation, other than the standard stepwise/rna/turner.wts (but if the score terms are different than in those turner.wts, then you need to modify the source code).

# Tutorial

See demos/public/recces/ for the latest demo for running RECCES.

# Options

Below are a list of available arguments for the recces_turner application.

-seq1 <String>
The sequence of the first strand (or the full sequence if it is single-stranded).

-seq2 <String>
The sequence of the second strand (skip if it is single-stranded).

-n_cycle <Int>
The number of Monte Carlo cycles.

-temps <Double/List of Doubles>
The simulation temperature. If it is a single value, the code performs
standard Monte Carlo at the given temperature. If the input is a list of
values, the code will run simulated tempering (ST).

-st_weights <List of Doubles>
The ST weights for simulations. Need to be specified if multiple temperatures
are given. Can be determined by short pre-runs (see demo).

-out_prefix <String>
Prefix for the RECCES output files.

-save_score_terms
If this option is supplied, RECCES will cache the values for each score terms.
Otherwise only the score histograms are returned.

-a_form_range <Double>
The sampling range for A-form conformations (duplex). Default is 60
(+/- 60 degress from ideal values).

-dump_pdb
If suppiled, the program will dump pdb files for examination.

-n_intermediate_dump <Int>
Dump the given amount of pdb structures for illustration purposes.

# Expected Outputs

Each RECCES run will generate one or several (depending whether ST is used) score histograms. If -save_score_terms is used, it also outputs the cached score terms for each sampled conformation. The result can then be analyzed using the Python scripts (see demo and tools/recces/README).