Information Technology Reference
In-Depth Information
Bioinformatics - Computational Support
for Genome Analysis
Fahri Salih KOCABAS
Middle East Technical University (METU) Computer Engineering Department
06530 Ankara, Turkey. (e-mail: fkocabas@tsk.mil.tr)
Abstract.
The major goal of bioinformatics is the analysis of sequence, structure and
function relationships. In these studies, lab experiments and computational work must
validate and consolidate each other, and findings of both initiatives expedite each other's
improvement. This process requires experts who can both work at lab bench and in
computer applications. This chapter summarises a computer scientist's views on the
diverse fields of bioinformatics.
Introduction
The work of bioinformatics requires to orchestrate different disciplines like molecular biology,
math, computer science, statistics etc. to have a united focus on its objectives in a team oriented
work environment. It is easy to state but difficult to implement. The existence of double major
scientists, appealing grants and the enthusiastic nature and the challenge of the subject may well
be organised and utilised to start and maintain such a bioinformatics study. Therefore, being a
computer scientist, the author values the information contained in this article even if most of the
content is known by related disciplines. It is so because the bringing the related information
together under the supervision and experience of a computer scientist is valuable. The major
concepts largely focused on sequence analysis are visited in the second part whereas the
concluding remarks and tips for future studies are given in the last part.
In gene and aminoacid sequence analyses, the sequences of related ones were observed to
be similar; thus, corresponding portions matched in their alignments. It is known that strong
similarity indicates the homology where homology means a common evolutionary history
whereas similarity emerges for some other criteria, not for a common ancestor [1]. Alignment by
utilising basic computer science techniques presents solutions to the question of the relatedness
of sequences. The genetic, functional and structural relations are under examination in this
regard.
Other than comparison analyses, the computational requirements of molecular biology
could mainly be listed as: set of tools powered by integrated knowledge bases; solid, complete
methodologies; computation techniques enriched with introduction of probability, uncertainty,
fuzziness, learning mechanisms, heuristics, approximation, knowledge discovery and the like.
One important aspect is to decide over the trade off between the sensitive, exact solution and
exponential computational running times. The bottom line is that the environment in which the
bioinformatics problem resides must well be reflected in designing the optimum data structures
and algorithms. It is also the main course of ongoing advances that bioinformatics graduate
professionals will employ bioinformatics - specific computational frameworks in line with the
advances in related disciplines in coming years.
