|
| -- BrandonGaytan? - 12 Sep 2008
- Ensure homepage has a photo and biography
- Two examples demonstrating how computational biology is relevant to synthetic biology
- Synthetic biologists attempting to engineer DNA sequences (e.g synthesizing a genome, perhaps Craig Venter style) must sequence their samples to demonstrate the validity of an experiment. Efficient computational tools, such as Velvet, can be used to quickly analyze and assemble large amounts of DNA sequence.
- Synthetic biologists attempting to recreate or adapt signal transduction pathways for different purposes (e.g. production of ethanol for biofuels) would benefit from examination of computational models of these regulatory networks prior to starting wet-lab experiments.
- The difference between a virus and transposon and their applications and limitations
- Differences: Transposons "jump" around the genome of a host cell, are unable to survive outside of the host cell, are inherited through the germline, replicate slowly (when the cell divides and replicates its DNA), and do not kill the host cell (this is not to the transposon's benefit- it cannot survive without its cell). In contrast, viruses, if they integrate at all (example HIV provirus), usually do not extract themselves and move to other parts of the genome. In addition, viruses can survive outside of a host cell, and are not usually passed through the germline. Moreover, viruses replicate very quickly by hijacking cellular machinery, usually causing rupture and death of the host cell, which releases viruses that can then infect other cells.
- Applications: The specific disruption of genes by transposons can offer insight into a gene's function. Viruses, such as the lentivirus, can be used to introduce new genes into an organism.
- Limitations: Methods that reliably deliver or disrupt DNA by viral vector or transposon are basically non-existent. While gene-therapy (introducing a missing gene or replacing a mutated gene to remedy a disease) may succeed a percentage of the time, it is dangerous to insert DNA into a human without 99.9999% reliable methods. DNA delivered by a virus could recombine in an undesired location in the genome, possibly disrupting a necessary functional gene and causing deleterious effects.
- Description of bioinformatics tools
- protein struture prediction
- Ab initio is a term used in the field of protein structure prediction to describe computational methods used to build protein structures from scratch or "_de novo_", rather than basing predictions on existing data. Although this field of study is growing, it utilizes vast computational resources and has only been used to predict the structure of tiny proteins. New algorithms, supercomputers, or distributed computing methods may all assist future ab initio efforts. An example is Touchstone II.
- Various comparative protein modelling softwares exist that utilize previously solved protein structures (usually through x-ray crystallography methods) as a template for the prediction of the folding/structure of an unknown protein sequence. There exist a limited number of tertiary structural motifs within proteins, so these softwares are very effective. The program MODELLER utilizes homology modelling, which is based on the assumption that two proteins that share sequence similarity will share structural similarity. If an unknown amino acid sequence is very similar to a solved sequence, this method would be preferable to ab initio techniques. Problems arise when the unknown amino acid sequence is very divergent from any solved structures, which makes the selection of and alignment of a template very difficult.
- gene function
- A controlled vocabulary facilitates database queries by creating consistent descriptions of search terms across different databases. For example, the vocabulary utilized by Gene Ontology helps "describe gene products in terms of their associated biological processes, cellular components and molecular functions in a species-independent manner" (reference), which allows for the creation of uniform queries across various databases. Issues include the linkage and integration of search terms and phrases. One may miss relevant database records if the search input does not match the vocabulary.
- Biochemical pathway databases, such as MetaCyc, provide a reference to biochemical pathways and enzymes, as well as support for metabolic engineering and computational prediction of metabolic networks. In contrast to a controlled vocabulary search (in which relevant topics could be missed due to errors in user input or the vocabulary), a query on one gene would display an entire metabolic network (the data is more integrated). However, inconsistencies in experimental data (or a complete lack of data) may retard the growth and integration of the database.
- structure analysis
- Protein/protein docking software, such as UCSF's DOCK, are used to model possible interactions and binding orientations between two proteins. These softwares can also be used to determine how/if mutations in the protein(s) affect binding interactions. Time and resources can be saved if experiments are carried out with docking software prior to possible wet-lab experiments. Issues arise when the proteins to be modelled are very large, as the time and memory required to solve the problem go up exponentially.
- Protein/small-molecule docking software is often built into protein/protein docking software (e.g. UCSF's DOCK), as the algorithms used are very similar if not alike. Docking a small molecule or ligand to a protein preferable to docking a protein to a protein, as the size of the small molecule/ligand facilitates and speeds the search for various binding orientations/locations on the protein. Often, if one has an idea of binding sites or important residues on a protein during protein/protein docking, it helps to model the interaction between the important sections of the proteins (this speeds up the process). Important DOCK applications (for drug design, drug targets, toxicology studies, etc) include databases searches for ligands/compounds that inhibit enzyme activity, bind a particular protein, or bind nucleic acid targets.
- sequence analysis
- Sequence assembly softwares, such as VELVET or AMOS, align short DNA sequences into one large sequence. These softwares are utilized during the sequencing of a genome. However, this assembly is not perfect- many important pieces of sequence may possibly be missing, sequence may be unrecognizable, and if one is not careful, contaminating DNA could be mixed in. A greedy algorithm is often used, which results in a suboptimal solution to the problem.
- Gene-finding software algorithmically identifies regions of DNA that are likely to contain protein encoding genes, RNA genes, or regulatory elements. An example is GENSCAN. This software is most often used during the annotation of a genome that has been sequenced using the aforementioned sequence assembly software.
- RNA structure
- RNA folding software predicts RNA secondary structure, such as hairpin or internal loops, from RNA sequence, much as protein prediction software predicts protein structure from amino acid sequence. An example is VIENNA.
- RNA design software attempts to do the opposite of RNA folding software. Essentially, it creates an RNA sequence based on a pre-defined secondary structure. An example is the RNA inverse package, which is included in the VIENNA software.
|
