This uses a sub-string search, so that "graph" matches both "graph" and "graphical".
This search only searches by keyword. If you'd like, you can also search by researcher.
Found 3 results.
Listing from newest to oldest
TR-CS-2005-26
Machine learning approaches to siRNA efficacy prediction.
Sahar Abubucker
RNA interference (RNAi) is being widely used to study gene expression and gene regulation via selective knockdown of gene expression, which is important for functional genomic studies. RNAi refers to the biological process by which short interfering RNA (siRNA) after incorporation into the RNA induced silencing complex (RISC) degrades complementary messenger RNA (mRNA) sequences. This knockdown of mRNA prevents it from producing amino acid sequences that are responsible for gene expression. Recent studies indicate that all siRNAs do not produce the same knockdown effects. Due to the high cost of synthesizing siRNAs and the extensive effort required to test siRNAs, rational siRNA design---a priori prediction of functionality for specific siRNAs---is a critical task for practical RNAi studies. Therefore, a key component of RNAi applications is the selection of effective siRNA sequences---ones that degrade at least 80% of the targeted mRNA. The goal of siRNA efficacy prediction is to aid in designing siRNA sequences that are highly efficient in degrading target mRNA sequences. Most of the current algorithms use positional features, energy characteristics, thermodynamic properties and secondary structure predictions to predict the knockdown efficacy of siRNAs. In this work, frequently occurring patterns in siRNAs are used to make predictions about their efficacy. Time after transfection is shown to be an important factor in siRNA efficacy prediction---a feature that has been ignored in previous efficacy studies. The relevance of these extracted patterns to previous results and their biological significance are analyzed. Random feature sets are generated and the ability of these sets to predict efficacy are studied and their results discussed. Our algorithm does not require any specialized hardware and consistently performs better than other publicly available efficacy prediction algorithms.
TR-CS-2005-23
A Practical Approach to Significance Assessment of siRNA Off-target
Effects in RNA Interference
Wenzhong Zhao and Terran Lane
Detection of potential cross-hybridization (or cross-reaction) between a short oligonucleotide sequence and a longer (unintended) sequence is crucial for many biological applications, such as selecting PCR primers, microarray nucleotide probes or short interfering RNAs (siRNAs). In this paper, we propose a flexible framework for estimating the significance of siRNA off-target effects on untargeted transcripts (messager RNA, or mRNA) in the RNA interference (RNAi) process. The framework can also be extended to other applications with minor changes.
We have developed and implemented a new homology sequence search framework -- siRNA Off-target Search (SOS). SOS uses a hybrid, q-gram based approach, combining two filtering techniques using overlapping and non-overlapping q-grams. Our approach considers three types of imperfect matches based on biological experiments, namely G:U wobbles, mismatches, and bulges. The three main improvements over existing methods are: the introduction of a more general cost model (an affine bulge cost model) for siRNA-mRNA off-target alignment; the use of separate searches for alignments with and without bulges, that enables efficient discovery of potential off-target candidates in the filtration phase; and the use of position-preserving and order-preserving hit-processing techniques, that further improves the filtration efficiency. Overall, SOS achieves better performance, in terms of speed and recall/precision, than BLAST in detecting potential siRNA off-targets.
TR-CS-2004-32
Efficient RNAi-Based Gene Family Knockdown via Set Cover Optimization
Wenzhong Zhao, M. Leigh Fanning and Terran Lane
RNA interference (RNAi) is a recently discovered genetic immune system with widespread therapeutic and genomic applications. In this paper, we address the problem of selecting an efficient set of initiator molecules (siRNAs) for RNAi-based gene family knockdown experiments. Our goal is to select a minimal set of siRNAs that (a) cover a targeted gene family or a specified subset of it, (b) do not cover any untargeted genes, and (c) are individually highly effective at inducing knockdown. We show that the problem of minimizing the number of siRNAs required to knock down a family of genes is NP-Hard via a reduction to the set cover problem. We also give a formal statement of a generalization of the basic problem that incorporates additional biological constraints and optimality criteria. We modify the classical branch-and-bound algorithm to include some of these biological criteria. We find that, in many typical cases, these constraints reduce the search space enough that we are able to compute exact minimal siRNA covers within reasonable time. For larger cases, we propose a probabilistic greedy algorithm for finding minimal siRNA covers efficiently. Our computational results on real biological data show that the probabilistic greedy algorithm produces siRNA covers as good as the branch-and-bound algorithm in most cases. Both algorithms return minimal siRNA covers with high predicted probability that the selected siRNAs will be effective at inducing knockdown. We also examine the role of "off-target" interactions - the constraint of avoiding covering untargeted genes can, in some cases, substantially increase the complexity of the resulting solution. Overall, however, we find that in many common cases, our approach significantly reduces the number of siRNAs required in gene family knockdown experiments, as compared to knocking down genes independently.
News & Events | People | Academics | Research | Jobs | Companies | Visitors | Computer Facilities | Home
