CPSC 545 - Project Ideas


General Remarks:


1. Collecting and Characterising RNA Molecules with Known Secondary Structure

Predicting the secondary structure of RNA molecules provides an important basis for understanding their function. Programs to predict such structures exist, such as mfold [1] and the fold function of the Vienna RNA package [3]. Also, a program to predict secondary structure of a pair of RNA molecules is available on the RNAsoft web page [5]. However, the prediction accuracy of these programs needs to be further improved, especially for long molecules.

The first stage of this project is to create a database of RNA sequences and their known secondary structure. The database would contain two types of such data: single RNA molecules and pairs of RNA molecules. The database should comprise biological molecules as well as molecules whose structures have been studied in vitro. Other information about each entry in the database may be needed: RNA type (tRNA, rRNA, snRNA, ribozyme, artificial, etc.), references, etc. Databases with single RNA sequence-structures already exist [6,7], but they are specialised on specific types of RNAs. Also, many more known pairs of RNA sequences and structures exist, but are spread out over the Internet. As regards pairs of RNA molecules, it is questionable whether any database exists, but many research papers give examples [8], which may be collected into the new database.

Assembling such a database would provide an important basis for improving our understanding of RNA secondary structure and may ultimately help to design new methods for better prediction of secondary structures.

The second stage of the project is to use a programme for statistically analysing collections of RNA structures, RNA Analyser, to investigate and characterise the properties of the RNA structures in the newly created database. RNA Analyser takes as its input a set of RNA secondary structures, and outputs statistics related to the number of multiloops, length of stems, size of internal loops, etc. The goal of this stage is to characterise the properties of various types of biological RNA structures. For this purpose, the existing RNA Analyser will likely have to be extended in various ways.

This project will involve a substantial amount of literature and web search. It will improve your research skills and your knowledge on RNA molecules.

References:

  1. Zuker's web site on RNA secondary structure prediction: bioinfo.math.rpi.edu/~zukerm/rna.
  2. For background on RNA secondary structure prediction, see the lecture notes at U. Washington: www.cs.washington.edu/education/courses/527/00wi
  3. I. Hofacker, ``RNA folding packages". 3 October, 2000. www.tbi.univie.ac.at/~ivo/RNA
  4. I.L. Hofacker, W. Fontana, P.F. Stadler, S. Bonhoffer, M. Tacker, and P. Schuster, ``Fast folding and comparison of RNA secondary structures.'' Monatshefte f. Chemie, Volume 125, 1994. pages 167-188. (Available from the above web site.)
  5. RNAsoft web page: www.rnasoft.ca.
  6. Gutell Lab Comparative RNA Web Site: www.rna.icmb.utexas.edu.
  7. M. Sprinzl, C. Horn, M. Brown, A. Ioudovitch and S. Steinberg, Compilation of tRNA sequences and sequences of tRNA genes, Nucl. Acids. Res. (1998) 26: 148-153.
  8. P. B. Rupert and A. R. Ferre-d'Amare, Crystal structure of a hairpin ribozyme - inhibitor complex with implications for catalysis, Nature (2001) 410.

2. Comparison of Empirical Potential Functions for Reduced Off-lattice Models of Protein Folding

The protein structure prediction problem is one of the most important problems in computational biology. Simplified models of protein folding have been widely used to study protein folding. Success in solving the protein structure prediction problem relies on the choice of an accurate potential energy function that can descriminate between native (functional) state of the protein and misfolded states. Reduced off-lattice models are models that are usually used for defining the configuration of the backbone of the protein: the protein main chain (consisting of C, O, N and C_alpha atoms) is modelled by means of two angles per residue (phi and psi). Side chains are modelled as units and the side chain propensities are taken into account by the energy potential function.

This project is to carry out an empirical comparison of at least two known energy potential functions of your choice (see references below) with respect to their accuracy in descriminating between native structures and misfolded states of given proteins using the standard RMSD measure of similarity between two structures (RMSD); furthermore, the correlation between the similarity of various structures of the same protein and the respective energy potential values should be studied.

References:


3. Protein Design in the 2D HP Model

De novo design of proteins is an approach for designing novel proteins with predetermined properties. It gives rise to the following protein design problem or inverse protein folding problem: Given a protein structure find an amino-acid sequence that folds into the specified structure as its native state. The two-dimensional Hydrophobic-Hydrophilic (2D HP) model is an extremely simplified model of protein structure that has been extensively studied by biochemists and physicists, and models certain properties of real proteins.

The project is to implement an algorithms for protein design in the 2D HP model (ideally, a new algorithm that has not been previously applied to the problem) and to carefully evaluate its performance against results known from the literature. Furthermore, given protein structures can be investigated in terms of "designability", e.g., by determining the number of sequences that have a given structure as their native structure. Implementations of high-performance algorithms for 2D HP protein structure prediction will be provided.

References:


4. Motif Finding Algorithm

In order to find potential sites in DNA sequences that play roles in control processes such as regulation of genes or splicing of introns, researchers use algorithms that can find certain types of "motifs" in DNA sequences.

Buhler and Tompa recently developed an interesting algorithm for finding certain types of planted motifs in sequences. Specifically, the problem they addressed is as follows (quoted from their paper):

"Suppose there is a fixed but unknown nucleotide sequence M (the motif) of length l. The problem is to determine M, given t nucleotide sequences each of length n, and each containing a planted variant of M. More precisely, each such planted variant is a substring that is M with exactly d point mutations."

In their conclusions they suggest that finding good algorithms based on other probabilistic models for planting motif variants, such as allowing insertions and deletions, would be valuable. How could this be done? Alternatively, suppose that the planted motif is present not in all sequences, but in a subset of them. Could you develop an algorithm that could predict such motifs?

Reference:


5. Multiple Sequence Alignment

The problem of finding alignments of multiple sequences (DNA or protein) arises in many bioinformatics applications and often forms the basis for further computations, e.g., in phylogenetic tree construction. Unfortunately, determining optimal multiple sequence alignments is computationally hard (under certain, not realistic conditions, the problem has been shown to be NP-hard). A variety of heuristic techniques have been applied to this problem, including Genetic Algorithms and Simulated Annealing.

The goal of this project is to implement and empirically study the behaviour of iterative refinement methods for multiple sequence alignment using the SP (sum of pairs) score.

After familiarising yourself with the general approach and conducting a literature survey, you might want to implement two or three different methods (these could be all rather closely related) and study their performance on a set of homologous biological sequences. You might also consider additional tests on sets of sequences obtained from applying simulated evolution (according to one of the commonly used sequence evolution models, e.g., the Jukes-Cantor model).

References:


6. Combinatorial DNA Design: A Coding Theory Approach

A DNA strand can be seen as a string over a quaternary alphabet with the letters A,C,G,T.

Combinatorial methods for designing DNA strands have been used to obtain sets of DNA strands that do not hybridize between themselves and only with perfect complements.

Some methods used for this purpose can be found in the coding theory literature [1,2,3]. Nevertheless, few publications report results on time complexity and algorithmic behavior for their methods. The project is to implement and possibly expand one (or more) of the algorithms presented in [1,2] and/or [3], and to characterise its performance, with the goal of gaining a better understanding of the empirical performance of these methods and possibly achieving improved results.

References:

  1. K.U. Koshnick (1991), "Some new constant weight codes," IEEE Transactions on Information Theory, 37, 370-371.
  2. G. Dueck and T. Scheuer (1990), "Threshold accepting: a general purpose optimization algorithm appearing superior to simulated annealing," Journal of Computational Physics, 90, 161-175
  3. F. Comellas and R. Roca (1993), "Using genetic algorithms to design constant weight codes," Proceedings of the International Workshop on Applications of Neural Networks to Telecomunication, LAwrence Erlbaum, Hillsdale, NJ, 119-124.
  4. A.E. Brouwer, J.B. Shearer, N.J.A. Sloane, W.D. Smith (1990), "A new table of constant weight codes," IEEE Transactions on Information Theory, 36, 1334-1380.

7. Design of DNA Strand Sets with Unequal Length

The accurate design of DNA molecules for biomolecular computations and nano-assemblies is of great interest to many researchers today. Many approaches to the problem of designing sets of DNA strand with desirable properties have been applied so far. Most of these methods, however, are designed to generate sets of strands that are all of equal length.

The project is to develop an algorithm (or eventually to modify a currently existing method [1]) to design DNA strands of different length. All strands must fullfil similar constraints (combinatorial and thermodynamic). The algorithm should be able to build new sets from scratch as well as to extend a initial set of DNA strands and/or to generate new DNA strands without modifying the existing ones, such that the entire resulting set fulfils all given constraints. The new algorithm should be analysed empirically and its performance should be compared to existing methods.

References:

  1. Dan C. Tulpan, Holger H. Hoos, Anne Condon, "Stochastic Local Search Algorithms for DNA Word Design," DNA 8 2002: 229-241.

last update 03/09/16, hh