Biology Reference
In-Depth Information
Thus all areas of the biological sciences have urgent needs for
the organized and accessible storage of biological data, power-
ful tools for analysis of those data, and robust, mathematically
based, data models. Collaborations between computer scientists
and biologists are necessary to develop information platforms
that accommodate the need for variation in the representation
of biological data, the distributed nature of the data acquisition
system, the variable demands placed on different data sets, and
the absence of adequate algorithms for data comparison, which
forms the basis of biological science.
89
In 1995, Michael Waterman published
Introduction to Computational
Biology: Maps, Sequences, and Genomes
, arguably the fi rst bioinfor-
matics textbook.
90
Waterman's topic was grounded in the treatment of
biology as a set of statistical problems: sequence alignment, database
searching, genome mapping, sequence assembly, RNA secondary struc-
ture, and evolution, are all treated as statistical problems.
91
In 1998, the fourteen-year-old journal
Computer Applications in the
Biosciences
changed its name to
Bioinformatics
. In one of the fi rst is-
sues, Russ Altman outlined the need for a specialized curriculum for
bioinformatics. Altman stressed that “bioinformatics is not simply a
proper subset of biology or computer science, but has a growing and
independent base of tenets that requires specifi c training not appropri-
ate for either biology or computer science alone.”
92
Apart from the ob-
vious foundational courses in molecular biology and computer science,
Altman recommended training for future bioinformaticians in statistics
(including probability theory, experimental statistical design and anal-
ysis, and stochastic processes) as well as several specialized domains
of computer science: optimization (expectation maximization, Monte
Carlo, simulated annealing, gradient-based methods), dynamic pro-
gramming, bounded search algorithms, cluster analysis, classifi cation,
neural networks, genetic algorithms, Bayesian inference, and stochastic
context-free grammars.
93
Almost all of these methods trace their origins
to statistics or physics or both.
94
Altman's editorial was inspired by the fact that solid training in bio-
informatics was hard to come by. The kinds of skills he pointed to were
in demand, and bioinformatics was the next hot career. Under the head-
line “Bioinformatics: Jobs Galore,”
Science Careers
reported in 2000
that “everyone is struggling to fi nd people with the bioinformatics skills
they need.”
95
The period 1997-2004 coincides with the second period
of rapid growth of bioinformatics publications (see fi gure 1.1). Between
