SVM- Implementation and algorithm design with codes for SiRNA selection
Introduction
There are few people in history who have had the opportunity to change the way future generations see the world. Although there is constant progress in the development of thought and knowledge in numerous disciplines of study, it is rare that something comes along that becomes the basis of all future study in the field.
One such idea was formed by Francis Crick, an English molecular biologist, physicist and neuro-scientist who contributed to research within the field of genetics. His collaboration with James D. Watson has led to the discovery of the structure of DNA. Their work with Maurice Wilkins also earned them a Nobel Prize for Physiology or Medicine “for their discoveries concerning the molecular structure of nucleic acids and its significance for information transfer in living material.” If that wasn't enough, Crick also is responsible for naming a now well-known cellular process, the “central dogma” of molecular biology.
The Central Dogma of Molecular Biology explains the unidirectional synthesis of proteins from DNA. DNA is first transcribed into RNA and then translated into proteins. Although at the time the hypothesis was deemed a dogma, the name seemed appropriate. As Crick explained, he, “simply applied it to a grand hypothesis that, however plausible, had little direct experimental support”. After decades of research within the field, the dogma still stands. The process of transcription and translation, as well as its subjects, DNA, RNA and proteins have been continually studied.
RNAi has become an important method for down-regulating gene functions. The active intermediate, siRNA, is increasingly used for functional studies in mammalian cells. One of the main obstacles of using siRNA as a tool is to find an "effective" target site on the mRNA.SVM provides researchers appropriate algorithmns for finding the best siRNA target sequences.
METHODOLOGY
The goal is to design an algorithm and implement it for siRNA selection using a Support Vector Machine. There are three main tasks in the process: define and build the necessary feature set, implement the Support Vector Machine, and evaluate the features. The following sections describe the steps necessary to achieve a working Support Vector Machine along with a useful presentation of the importance of the features independently and when implemented together.
RNA interference (RNAi) is widely studied for its importance in genomic studies and its potential use in therapeutics. It is the mechanism by which messenger RNA (mRNA) sequences are degraded by complementary short interfering RNAs (siRNAs) incorporated into RNA induced silencing complex (RISC).
Recent studies indicate that all siRNAs do not produce equal knockdown effects. In vivo experiments to observe siRNA functionality are expensive and time-consuming. Different studies have proposed several characteristics
of the siRNA that indicate functionality, including the presence or absence of certain nucleotide in certain positions in the siRNA, thermodynamic properties related to stability and secondary structure.
We propose a computational approach to siRNA efficacy prediction that makes use of frequently occurring positional patterns in the siRNA data to discriminate between functional and non-functional siRNAs.
Requirements for RNAi in Mammalian Cells
Two or more effective siRNA(s) per target.
Efficient, reproducible delivery method.
Assays to determine siRNA effectiveness and RNAi effect.
Positive and negative siRNA controls.
Need for Effective siRNA
Published design guidelines alone are ineffective: Only 50% of the siRNAs give more than 50% silencing Only 25% of the siRNAs give more than 70% silencing. Generally testing four or five siRNAs Leads to high cost and invested time for researcher.
Benefits of High Efficacy SiRNAs
Use less siRNA per transfection and.
Minimize off-target effects.
Reduce toxicity.
Reduce costs.
How to Get Effective SiRNAs
Design.
Synthesize
Lists
Validate.
Aspects of siRNA Experiments
Two or more effective siRNA(s) per target
Efficient, reproducible delivery method
Assays to determine siRNA effectiveness and RNAi effect
Positive and negative siRNA controls
eed a sophisticated a
We assumed that the time after transfection is an important feature in determining efficacy.
Feature sets consisting of random positional patterns in the data will show reasonable performance leading to the hypothesis that the positional features used for predicting efficacy could be a consequence of limited data...
To be Continued in Next Blog.................
