SEARCH

SEARCH BY CITATION

Keywords:

  • Bioinformatics;
  • course design;
  • interdisciplinary curriculum;
  • problem-based learning

Abstract

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

We describe an innovative bioinformatics course developed under grants from the National Science Foundation and the California State University Program in Research and Education in Biotechnology for undergraduate biology students. The project has been part of a continuing effort to offer students classroom experiences focused on principles and reasoning, using a computer-assisted, problem-based learning model. Here we report on the course design, methods of assessment for the course and materials, and analysis of results obtained in initial offerings of the course.

In their 2003 report, BIO 2010: Transforming Undergraduate Education for Future Research Biologists [1], a committee of the National Research Council charted a course for educating 21st century biological scientists. The ambitious proposal, aimed at implementation by the year 2010, envisions interdisciplinary undergraduate experiences that seamlessly integrate biology with mathematics, statistics, computer science, physics, and chemistry. The research community has enthusiastically endorsed this new, quantitative bioscience curriculum, viewing the proposed restructuring as essential to transcend the traditional barriers between biology and mathematics-based sciences [2]. The changes, they say, will benefit students in mathematics and the physical sciences by providing biological systems to model while preparing biology students for the new world of quantitative biology. The relatively new discipline known as bioinformatics already provides a bridge from molecular and structural biology to computer science, statistics, and information theory. Adding a bioinformatics course to the undergraduate biology curriculum thus represents an easy, natural step toward the goal of providing quantitative, interdisciplinary coursework for all bioscience students.

DESIGN ISSUES

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

The Central Concept —

For students, understanding the fundamental concept of a course can illuminate every aspect of their learning experience. A central, powerful concept provides a framework for thinking, not only through the topics explored in the class but also through related courses and experiences in the real world. For course designers, the explanatory power of a fundamental concept provides an organizing principle. Construction of a course around a primary concept increases the likelihood that students will take away the essence of the discipline and a usable set of thinking skills [3].

At its core, bioinformatics is the extraction of biological information (or meaning) from biological data stored in databases. All other topics and questions in bioinformatics, such as how biological data are collected and stored, how protein structure and function can be predicted from sequence data, or how metabolic networks may be visualized, can be understood in relation to this primary concept. The overall design of our undergraduate bioinformatics course reflects our effort to keep this fundamental idea in focus.

Curricular Goals—

The curriculum project aimed to establish a foundation course in bioinformatics for third-year undergraduates at Sonoma State University, articulated with the biological science major and available as an elective for other majors in the School of Science and Technology. The plan called for us first to assemble, implement, and field test a computer-based module for diagnostic pretesting and remediation for enrolling students. The second task was to develop a set of problem-based instructional units to be used in a discussion/laboratory format with laptop computers and web browsers. Once these elements were in place, the focus shifted to implementation and evaluation of an introductory course in bioinformatics using these materials. Having run the course in successive semesters, the final task is to share the results of this effort with faculty at other undergraduate institutions.

Two courses offered in the biology program, biometry and ecology, currently require introductory statistics to be taken in the mathematics department. Department offerings with substantial population genetics content, such as systematics and evolution, emphasize quantitative reasoning without requiring specific mathematics courses. The bioinformatics course is the first to offer significant computer science content in the context of biological problems.

CONTENT AND ORGANIZATION

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Bioinformatics Content—

Bioinformatics has matured into a discipline by drawing on natural science disciplines (biology, chemistry, physics), quantitative disciplines (computer science, mathematics, statistics), and philosophy (ethics, philosophy of science). An important part of our task as educators has been to define specific areas of knowledge needed to construct an undergraduate bioinformatics curriculum. From biology, we have included molecular biology, genetics, cell biology, and evolution. Computer science has contributed programming, data structures, algorithms, databases, software engineering, and information theory. Relevant subject matter within statistics includes probability theory, experimental statistical design and analysis, and stochastic processes. Philosophy has given ethics as a framework to consider the societal impacts of technology, privacy and security issues, and standards of scientific conduct. Another subdiscipline of philosophy, the philosophy of science, is key to making sense of technological change, paradigm shift, scientific reasoning, proof and validation, and scientific accountability.

Graduate courses in bioinformatics generally include genome analysis, sequence analysis, gene expression, systems biology, data and text mining, phylogenetics, genetics and population analysis, databases and ontologies, and structural bioinformatics. In Spring 2002, a group of California State University science faculty met to create content guidelines for undergraduate bioinformatics courses in the California State University system, intended to bring these consensus areas within the reach of undergraduate students. The resulting set of recommendations provided the initial list of topics for our course. After some experimentation, we arrived at the syllabus shown in Table I. All materials developed for the course have been posted on a web site, available at www.bscbioinformatics.com/Stu/.

Objectives for Target Knowledge and Skills—

With respect to skills, experts at the undergraduate level generally agree that student quantitative and analytical capabilities must be developed in four areas: effective use of information technology to access data resources, problem solving using mathematical tools, statistical analysis, and modeling of biological systems [4]. Integral to the course design process was the development of a set of bioinformatics learning outcomes based on consensus content (see above) and skill areas. Objectives were therefore established at the outset for subject-matter knowledge, discipline-related skills, problem solving, critical thinking, and attitude development. (Fig. 1)

Organization of Course and Materials—

A major concern in designing a new elective course, especially one with interdisciplinary content, is the entering competencies of the students. Each enrolling student deserves a fair chance to succeed, since each takes a risk when attempting a potentially difficult course. Inclusion of a pretest unit in the course was our attempt to create a level playing field. The diagnostic pretesting unit anticipates the background needed for success in the course. Here each student evaluates his or her preparation through self-testing. A selection of tutorial materials is available for review by the students based on their self-assessment.

To design the unit, it was first necessary to define the prerequisite knowledge and skill set. We estimated the level of knowledge in mathematics (including statistics and probability), computing, biochemistry, molecular biology, and evolutionary genetics needed to successfully navigate the course. This estimate was based on pretest data on the student population likely to enroll in the course, together with our previous experience with California State University science students. Standards were then specified for remedial materials. Specifications for the tutorial materials were arrived at after an extensive review of educational web sites. We selected six categories of tutorials, primarily on the basis of their relevance to the course content, for the quality of the information presented, and for their user interface. We tried to find at least two good tutorials in each subject area, both to mitigate the broken link problem and to accommodate different learning styles. We then designed a 30-question diagnostic pretest implemented in JavaScript for display in an HTML browser [5]. After looking at various methods for creating a structure for organizing and linking review materials to the pretest, we opted to post the pretest and a page containing the list of hyperlinked tutorials on the course web site.

The first week of each course offering is devoted to diagnostic testing of enrolling students. We ask each student to fill out an answer sheet on the first run through the test, which we collect as part of our assessment. Each question on the answer sheet includes a confidence assessment so that the student may indicate his or her level of confidence in the answer. Based on the self-scored online quiz, students needing to acquire background concepts or skills are encouraged to proceed to the tutorials and then retake the quiz. Students needing little or no remediation are offered the opportunity to tutor their student colleagues or act as consultants on the quality and effectiveness of the tutorials. The process for the first week of the course is shown in Fig. 2.

The main part of the course consists of classroom discussions and computer laboratory time blocks interspersed with time for group work and online activities. The computer laboratory time is used for online exercises, directed activities, problems, projects, discussions, and student presentations. As shown in Table I, the bioinformatics content is divided into six instructional units, each 2 weeks in length, except for the first, which was extended by a week to include an introduction to computing. The units cover the consensus content areas of bioinformatics. Unit 1 focuses on databases, ontologies, and text mining. Unit 2 takes up genome analysis. Unit 3 explores gene expression and genetics. Unit 4 is devoted to phylogenetics and molecular evolution. Unit 5 introduces students to structural bioinformatics. Unit 6 attempts to integrate concepts learned in the previous units through a survey of systems biology.

A thematic problem or project is assigned in each unit. Most of these are contributed by colleagues in the School of Science and Technology based on their own research interests. These faculty present a problem to the students, giving them sufficient background to get started, and then return at the end of the unit to discuss the results with the class.

ADAPTATION AND IMPLEMENTATION

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Online Course Materials—

Instructional materials were collected from a variety of sources or were created by the instructors for posting on the course web site [6]. The materials cover topic areas across the scope of bioinformatics, including background information on each topic, text readings, hyperlinked exercises with links to database and application resources, tutorials, manuals, guides, published articles, problems, projects, and directed activities. The textbook used for the first two offerings of the course clearly presents background on computation, structure, and access to databases, as well as providing a nice introduction to a wide range of bioinformatics software [7]. The web site has a number of features in addition to the unit materials, including a glossary, a calendar, external links, a discussion forum, a library of articles in PDF format (password-protected), tutorial collection, self-tests, and public domain software for download.

Problems and Projects—

Each unit of instruction contains instructional materials fitting one of four categories. Projects are activities in which students collaboratively generate and analyze results. Problems require students to collaboratively define issues, investigate, and devise solutions based on the University of Delaware Problem-Based Learning model [8]. Directed activities guide students through the solution to a problem, given an approach and a set of resources. Exercises are designed by the instructors to build individual skills and concepts.

Featured Software Packages—

Rather than surveying the full array of applications available in the topic area, we have chosen to concentrate on one widely used package. Each unit thus features one important software package. Students explore an application in each category in some depth, gaining experience by using the software in exercises and projects and investigating how the software functions in an algorithmic sense. We have applied the following criteria in choosing bioinformatics programs for use in the course. 1) The application must be widely used, with published performance benchmarks versus other packages. 2) The software must be in the public domain (i.e., open source). 3) A published description must be available in the literature, preferably as a formal research article. 4) The source code must be available for download and local installation. 5) The application must be freely available online through an application service provider, preferably with a good tutorial, so that students can familiarize themselves with its capabilities and features.

Instruction includes reading the original literature on the software tool, introduction to the algorithmic approaches underlying the tool's design, plus concrete examples of how the software is used. For example, we introduce BLAST [9] by directing students to a BLAST application service provider, the National Center for Biotechnology Information (NCBI)11 [10]. In a series of laboratory activities, students step through the process of using DNA and protein sequences as data base queries. They investigate the ever increasing array of databases and software available at this national resource [11]. Having seen what BLAST can do, students go through the tutorial on the statistical principles underlying BLAST's scoring function [12], answering study questions for homework. The class is assigned to read and discuss the original paper [9] as a problem-based learning activity [13].

Each unit problem attempts to develop student understanding of the bioinformatics concepts associated with the application domain of a category of bioinformatics software. For BLAST, group projects are assigned that require students to use BLAST and online data resources to solve a research problem. In general, we want students to know what the applications do, how they work, what kinds of problems each one is designed for, where to get them, and how to get them running.

ASSESSMENT

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Philosophy—

Formative assessment can guide teaching and learning throughout a course [14], making this approach particularly attractive during development of a new offering. To be effective, formative evaluations must be embedded in the course design so that they are integral to the teaching and learning process. Assessments contribute most to learning when done continuously by instructors and students working together and when they are firmly grounded in student learning objectives. In a course where the instructors are learning as much as the students, formative assessments provide critical, real-time information on what is (or is not) working.

Methods—

Unit objectives were written with the following criteria in mind. (a) Instruction must reinforce critical thinking, written and oral communication, and problem-solving skills. (b) Students must be given opportunities to apply knowledge and skills to solve realistic problems. (c) Learning must be related to the real world. (d) Instruction must emphasize active, experiential learning. (e) Objectives must allow assessment to be based on student performance and products. (f) Collaboration must be encouraged. The learning outcomes for the course are listed in Fig. 1. Because we found it impractical to meet all of these criteria in a single unit, the nature of each unit's content determined which subset of the learning outcomes in Fig. 1 should be incorporated.

Evaluation of the curriculum project as a whole required a separate set of criteria. Here we needed to ask how effectively the course engaged student interest in bioinformatics as a discipline, whether student understanding of bioinformatics was enhanced, and the extent to which each unit succeeded in highlighting the algorithmic underpinnings and limitations of featured software applications. Evaluation of the course focused primarily on its organization, the quality and effectiveness of the course materials, and the effectiveness of the active learning pedagogy.

Given the learner-centered approach and problem-based learning model, assessments emphasized performance (i.e., the quality of skills and products). We tried to make them authentic (i.e., reflective of real-world tasks) whenever possible. Performance assessments included, for example, presentations and project reports like those given in a university research group. All student assessments developed or adapted for the course were criterion-referenced rather than comparative. We sought to evaluate each student on his or her individual achievement of standards, rather than emphasizing comparison with other students. Rubrics were developed for scoring student performance on exercises, graded unit study questions, a mid-semester problem set, projects, problems, and directed activities [15]. These rubrics rewarded mastery of concepts, the ability to think through problems, presentation skills, written work meeting a negotiated standard of quality, and evidence of desirable attitudes [14].

Using a variety of assessment methods not only allows for cross-validation of the results but can also reveal underlying trends not obvious from any individual type of analysis [16, 17]. We therefore collected and analyzed data from student feedback on surveys, from criterion-referenced grading of assignments, from traditional examinations, from institutional teaching evaluations, and from standardized testing. Student assistants conducted usability surveys of the course materials by observing in-class activities and by personally interacting with the materials. Data were collected on the usefulness of the materials, how long the assignments took to perform, and how well their peers responded to the combination of materials and instruction.

Student self-assessment has many advantages in an active learning environment, not the least of which is the development of the critical thinking, ethical, and analytical skills listed in the learning outcomes [18]. It is especially important in an interdisciplinary course for students to take responsibility for their own learning. In that spirit, the bioinformatics students assessed their own learning in relation to the materials and the instruction provided in the course. Six unit surveys were solicited from each student in the course, one at the end of each instructional unit.

Active, problem-based learning requires a different set of learning skills, which are sometimes referred to as higher-order thinking skills or deep learning approaches [19, 20]. A few diagnostic tools are available for assessment of college student epistemology. We adapted Biggs' Study Process Questionnaire [19] to evaluate the motivations and strategies of students enrolling in the bioinformatics course. A pilot study using this tool was done on one group of students, who completed Biggs' questionnaire as part of the diagnostic unit.

RESULTS

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Pretest Unit Results—

Table II summarizes the results of the 30-question pretest. The test covers six subject areas: biochemistry, computing, evolutionary genetics, web browsing, probability and statistics, and molecular biology. The questions (five per subject) focus on basic principles and applications expected to be within the general knowledge of computer-literate students having taken at least one course in molecular biology. Because the number of students is small, no claims can be made for statistical significance, but the trends can provide food for thought.

We were surprised that more than 80% of the students enrolling in the course could not correctly answer more than three of five questions about evolutionary genetics, especially given that most of the students were biological science majors in a program with a strong emphasis on macro evolution. Performance on the chemistry, math, and molecular biology questions was somewhat better. Only 50–60% missed two or more questions in these areas. The level of background knowledge of computing and web-related concepts was higher than expected. More than two-thirds of the enrolling students could correctly answer at least three of the five questions in those categories (Table II).

It was of interest to look for correlation between a lack of prerequisite knowledge and the probability that a student would drop the course. The probability of dropping was estimated from the number of students known to have dropped divided by the total number of students whose sub scores were in a given range. Counterintuitively, the data show that lack of computing knowledge or web savvy was not correlated with a decision to drop out of the course (Table II). The same percentage of students stayed in the course whether they scored below the threshold of 3 or less or above the threshold. However, lower scores in the chemistry and molecular biology sections did appear to correlate with increased likelihood of withdrawal from the course. Only 24% of students scoring better on the chemistry section dropped out, whereas the withdrawal rate was 36% among those less prepared in chemistry. Fully half of those less prepared in molecular biology chose to abandon the course.

Results from Surveys and Performance Assessments—

We devised a questionnaire to ask direct questions about what the students believed were the main ideas they had learned in each unit and which elements of the unit had been most helpful. To collect and analyze the responses, we created a database application with a simple input form matched to the questionnaire. Qualitative analysis of the unit surveys revealed very positive student reactions to the course materials. For all six units, students reported having learned at least one new concept from exercises designed by the instructors for individual skill and concept building. Directed activities (in which students are guided through the solution to a problem when given an approach and a set of resources) also appear to have been effective vehicles for conveying key concepts in bioinformatics. Students uniformly enjoyed the carefully structured unit projects in which they collaboratively generated and analyzed results. They were divided in their enthusiasm for the open-ended problems that required them to define issues, investigate on their own, and devise solutions based on independent analysis.

Interestingly, the questionnaire revealed widely varying detail and complexity in student descriptions of concepts they had learned. Three responses to the question about concepts learned in Unit 4 (Phylogenetic Analysis) illustrate this range. Student A answered that he learned “how to generate a phylogenetic tree.” Student B answered that he learned “assembly of phylogenetic trees using DNAPARS (a component of PHYLIP), ClustalDist, Drawtree, and Drawgram.” Student C answered that he learned how to “establish phylogenetic relationships and determine rates of evolution along lineages.” Although each student clearly got the target concept, these responses show depths of understanding ranging from superficial to profound.

Results from the Study Process Questionnaire—

Biggs' analysis identifies three main motivations for learning: superficial (motivated to minimally get by), deeply interested (motivated by curiosity, desire to understand), and achievement-driven (goal-oriented, desire to excel) [19]. Three general study strategies emerge from Biggs' work: superficial strategies (minimalistic, rote learning), deep learning (wide reading, making connections), and achieving strategies (being organized, scheduled, and disciplined). The Study Process Questionnaire (SPQ) uses 42 questions to evaluate a student's motives for learning and the strategies he or she generally uses. Comparison of a student's scores to published norms yields the student's individual motivation and strategy profile. Some combinations of motivation and strategy are problematic, but if identified early, they can be improved with appropriate instruction and counseling. Although nine combinations of motive and strategy are possible, in practice, students tend to take one of four approaches to their academic work: superficial, deep, achieving, or deep achieving.

We gave this test to 10 enrolling bioinformatics students during the first week of the semester. Analysis of student SPQ profiles (data not shown) found a remarkable 7 out of 10 of these students harboring a deep learning motivation, either predominantly or exclusively. This result correlated well with survey responses in which most of the entering students expressed a strong interest in the subject matter with great curiosity about bioinformatics. The result further suggests that the flexible structure of the class, opportunity for independent exploration, and criterion-referenced grading were a very good fit to the learning styles of the student audience.

Unexpectedly, the profiles also revealed that six of these students had adopted a low achievement strategy, characterized by poor study habits, inefficient time management, and disorganization! That a majority of the incoming class had adopted a low achieving study strategy (on Biggs' scale) correlated well with our subsequent observation that the students were not using their class time effectively. The instructors were continuously frustrated by the students' tendency to procrastinate rather than start assignments right away. The assessment team found itself defining two types of procrastination, which we called Type I (student knows how, just won't) and Type II (student wants to, just can't). From our tiny experiment, it appears that both types of procrastinators can be identified by their SPQ profiles. Type I students show up as having low achievement motivation, whereas Type II students are low on the achieving-strategy scale [19].

The SPQ may provide key insights into the reasons why students might not be getting as much out of the course as we hope and expect. It is often difficult to gauge whether a lack of student performance is due to excessive content, a pace that is too quick, or concepts that are too difficult for the student audience. Science teachers constantly worry about the delicate balance between challenging and frustrating the students in a course [20]. Frequently, the expedient solution is to lower the level of the material presented. The SPQ may give us a way to diagnose student performance problems, take appropriate steps, and ultimately, increase the quality of the course content.

DISCUSSION

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

Private sector demand for people with computational molecular biology skills drove the early development of bioinformatics courses and curricula [21]. The perception that such skills would ensure employment opportunities in big pharma and biotechnology led a number of institutions to start bioinformatics ”training“ programs. Students can earn certificates, bachelor's and master's degrees, and doctorates in bioinformatics. Recent trends suggest, however, that the bioinformatics labor market has not expanded as forecast; in fact, private sector demand has shrunk [22]. Bioinformatics courses and curricula can no longer be justified solely on vocational grounds.

We see the emerging discipline of bioinformatics not as a focus for training but rather as an opportunity to introduce quantitative, computational molecular biology into the undergraduate curriculum. Typical bioinformatics problems provide a rich source of interdisciplinary experiences in which students must combine biological knowledge with computer science, modeling, probabilistic analysis, and so forth. In this way, undergraduate bioinformatics courses represent a step toward the goal of a more quantitative biological science curriculum envisioned by the authors of BIO 2010 [1].

thumbnail image

Figure FIGURE 1.. Bioinformatics learning outcomes.

Download figure to PowerPoint

thumbnail image

Figure FIGURE 2.. Pretesting and review unit.

Download figure to PowerPoint

Table Table I. Bioinformatics course content
UnitNameTopics (alphabetical)Featured software
SNP, single-nucleotide polymorphism; PSSM, position-specific scoring matrix; HMM, hidden Markov model.
1DatabasesData, sequence and text miningBLAST
  Database integration 
  Design, implementation of biological databases 
  Ontologies 
  Query design 
  Sequence alignment 
2GenomicsAnnotationBIOPERL
  Comparative genomics 
  Genome analysis 
  Genome projects 
  Model organisms 
  Sequence analysis, fragment assembly 
3Molecular GeneticsAnalysis of gene expression dataMEDMINER
  Disease genes 
  Gene expression (central dogma) 
  Genetics and population analysis 
  Microarray technology 
  SNP and haplotype mapping 
4PhylogeneticBayesian inferencePHYLIP
     AnalysisDistance matrices 
  Molecular evolution 
  Multiple sequence alignment 
  Phylogeny analysis 
  Tree-building methods 
5Protein StructureClassification, folds and motifsPSI-BLAST
  Folding, dynamics, secondary structure prediction 
  Homology, threading, and ab initio modeling 
  Protein structure, function, and evolution 
  Profile methods (PSSMs, HMMs) 
  Structure determination (X-ray, NMR) 
6Metabolism andCluster analysisCLUSTER
     NetworksMetabolic pathways 
  Modeling complex data 
  Pharmacogenomics 
  Protein interaction networks 
  Systems biology overview 
Table Table II. Pretest results for 23 students enrolling in introductory bioinformatics
Subject areaMean scoreStandard deviation
   % enrolling% dropped% enrolling% dropped
Chemistry3.31.061 (14/23)36 (5/14)74 (17/23)24 (4/17)
Computing3.71.235 (8/23)25 (2/8)78 (18/23)28 (5/18)
Evolution2.41.087 (20/23)35 (7/20)61 (14/23)36 (5/14)
Web savvy3.81.226 (6/23)33 (2/6)78 (18/23)33 (6/18)
Mathematics3.11.061 (14/23)29 (4/14)70 (16/23)31 (5/16)
Molecular biology3.51.352 (12/23)50 (6/12)70 (16/23)19 (3/16)

Acknowledgements

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES

We thank colleagues Richard Whitkus, Judy Sakanari, Derek Girman, and Balasubramanian Ravikumar for their active participation in the development and delivery of the course. Comments on the manuscript by Kathleen Fisher, Director of the Center for Research in Mathematics and Science Education at San Diego State University, were much appreciated.

  • 1

    The abbreviations used are: NCBI, National Center for Biotechnology Information; SPQ, Study Process Questionnaire.

REFERENCES

  1. Top of page
  2. Abstract
  3. DESIGN ISSUES
  4. CONTENT AND ORGANIZATION
  5. ADAPTATION AND IMPLEMENTATION
  6. ASSESSMENT
  7. RESULTS
  8. DISCUSSION
  9. Acknowledgements
  10. REFERENCES
  • 1
    L. Stryer, R. Breslow, J. Gentile, D. Hillis, J. Hopfield, N. Kopell, S. Long, E. Penhoet, J. Steitz, C. Stevens, S. Ward (2003) BIO 2010: Transforming Undergraduate Education for Future Research Biologists, National Research Council, National Academies Press, Washington, D C.
  • 2
    W. Bialek, D. Botstein (2004) Introductory science and mathematics education for 21st-century biologists, Science 303, 788790.
  • 3
    G. Nosich (2005) Teaching Students Fundamental and Powerful Concepts, 25th International Conference on Critical Thinking, July 9, 2005, Berkeley, CA.
  • 4
    C. Hack, G. Kendall (2005) Bioinformatics: Current practice and future challenges for life science education, Biochem. Mol. Biol. Educ. 33, 8285.
  • 5
    R. M. Horton (1999) An introduction to Perl: The QuizMaker revisited, BioTechniques 27, 470472.
  • 6
    Bioinformatics Teaching and Learning (2005) Bioinformatics Course Materials Home, www.bscbioinformatics.com/Stu.
  • 7
    C. Gibas, P. Jambeck (2001) Developing Bioinformatics Computer Skills, O'Reilly & Associates, Inc., Sebastopol, CA.
  • 8
    B. Duch, S. E. Groh, D. E. Allen (2001) The Power of Problem-Based Learning, Stylus, Sterling, VA.
  • 9
    S. F. Altschul, W. Gish, W. Miller, E. W. Myers, D. J. Lipman (1990) Basic local alignment search tool, J. Mol. Biol. 215, 403410.
  • 10
    NCBI (2005) NCBI BLAST — The Basic Local Alignment Search Tool, www.ncbi.nlm.nih.gov/BLAST.
  • 11
    NCBI (2005) BLAST Program Selection Guide, www.ncbi.nlm.nih.gov/BLAST/producttable.shtml.
  • 12
    NCBI (2005) The Statistics of Sequence Similarity Scores, www.ncbi.nlm.nih.gov/BLAST/tutorial/Altschul-1.html.
  • 13
    H. B. White (1996) Addressing content in problem-based courses: The learning issue matrix, Biochemical Educ. 24, 4145.
  • 14
    J. M. Atkin, P. Black, J. Coffey (2001) Classroom Assessment and the National Science Education Standards, National Research Council, National Academies Press, Washington, D C.
  • 15
    B. G. Davis (1993) Tools for Teaching, Jossey-Bass Publishers, San Francisco.
  • 16
    K. P. Cross, M. H. Steadman (1996) Classroom Research: Implementing the Scholarship of Teaching, Jossey-Bass Publishers, San Francisco.
  • 17
    B. E. Walvoord, V. J. Anderson (1998) Effective Grading: A Tool for Learning and Assessment, Jossey-Bass Publishers, San Francisco.
  • 18
    B. D. Wright, in J.Gaff, J.Ratcliff, Eds. (1997) Handbook of the Undergraduate Curriculum, Jossey-Bass Publishers, San Francisco.
  • 19
    J. Biggs (1987) Student Approaches to Learning and Studying: Study Process Questionaire Manual, Australian Council for Education Research, Melbourne.
  • 20
    B. S. Chapman (2001) Emphasizing concepts and reasoning skills in introductory college molecular cell biology, Int. J. Science Educ. 23, 11571176.
  • 21
    G. C. Black, P. E. Stephan (2004) Bioinformatics: Recent Trends in Programs, Placements and Job Opportunities, Report No. 2002-5-58 BCMB, Alfred P. Sloan Foundation, New York, NY.
  • 22
    G. C. Black, P. E. Stephan (2005) Bioinformatics training programs are hot but the labor market is not, Biochem. Mol. Biol. Educ. 33, pp. 5862.