iology is not like physics. Physicists are taught to analyze everything down to irreducible components (molecules to atoms to protons to quarks to . . .). Early on, biologists learn to cope with the irreducible complexity of living things. For several years now, we have been creating and experimenting with software that embodies a physicist's conception of biology.

GenScope⁵™ is a computer-based manipulative (CBM) - a manipulable model that we have created to help students learn genetics. We described the program last month (@CONCORD, Winter 1998¹), and you can learn more about it by visiting our web site. Suffice it to say that GenScope cuts a long but narrow swath through the biology curriculum, touching on DNA, chromosomes, cells, organisms, pedigrees, and populations, but viewing each through the restricted focus of genetics. For example, cells in GenScope are entirely undifferentiated and contain nothing more than chromosomes. Their only behaviors are mitosis and meiosis, and their only function is to carry genetic information from one generation to the next.

But biology does not suffer gladly such compartmentalization. Seemingly straightforward questions - e.g., "Is cancer a genetic disease?" - cannot be answered by reference to genetics alone, but may entail a discussion of the life cycle of retroviruses, the effect of environmental factors on DNA, the role of DNA in protein formation, the function of proteins in regulating cell division and death, and metastasis and histologic influences on cells. Biology is integrative, not reductionist. What is to be done? We are pleased to announce the imminent appearance of a new CBM, to be named BioLogica™™, that will address the mismatch between the narrow focus of GenScope and the wide-ranging nature of BioLogica™l reasoning. BioLogica™ will expand significantly on its illustrious predecessor, both in content and, we trust, in pedagogical power.

Molecules and Cells

BioLogica™ is smart. It knows about molecules beyond DNA; it knows that cells contain many wondrous things besides chromosomes; it knows about relations between cells. Actually, it doesn't really "know" any of these things, of course, but it has models of them and, like GenScope's, the models are manipulable by the student, the results of manipulations of any one model being instantly reported to all the others. Thus, BioLogica™ imitates life in being essentially a collection of semi-autonomous entities (molecules, cells, tissues, organisms) that interact through the exchange of information. This makes BioLogica™ quite complex.

Leaving aside a certain undeniable technological hubris, why do we need such a complicated program? Our ultimate goal is to help students think like biologists. Any computer program forces its users to think in ways that reflect the way the software is constructed. (To the person whose only tool is a spreadsheet, every problem looks like a grid.) By encompassing the complex interactions that characterize living things, BioLogica™ will give students a BioLogica™l "sandbox" within which to hone their reasoning about living things. Here is an example.

Let's go back to that seemingly simple question about cancer. Imagine that BioLogica™ exists, and we have challenged a student, or more likely a pair of students, with a puzzle. We have given them an organism - perhaps a fictitious one like the GenScope dragons, perhaps a simplified model of a real one - that harbors a colony of cancerous cells. The problem for the students is to locate these cells, figure out what has gone wrong with them, and if possible fix it.

Cancerous cells generally look different from normal ones under a microscope. BioLogica™ will model this difference, probably by linking to photographs of real cells, so by comparing these the students should be able to find the "bad actors." BioLogica™ will also show cells as they appear in textbooks, with some enhancements. Animated diagrams will depict the flow of critical factors from one organelle to another or across the cell membrane. A different view will highlight features of the cytoskeleton, the complex network of protein filaments that extends throughout the cytoplasm. Yet another will display the life cycle of the cell. This latter feature will demonstrate to the students that the cancerous cells are continuing to divide long after the normal ones have become quiescent. The question is, why?

For pedagogical purposes we may have to simplify matters somewhat, but BioLogica™ will faithfully model the basic concept of a cell cycle control system mediated by molecular reactions. We will offer students an array of tools with which to measure the flows and concentrations of different proteins. Some of these proteins enable the cell to divide. In normally functioning cells these proteins become inactivated under certain conditions - e.g., in response to overcrowding from neighboring cells - and the cell essentially "hibernates." However, when the gene that codes for it is mutated the result can be a failure of the protein to inactivate, resulting in uncontrolled proliferation (mitosis) of the affected cell and the growth of a tumor.

In order to "cure" their organism's cancer, students must first compare its cancerous cells to normal ones. They will find that in the normal cell certain reactions fail to occur because of the absence of a "growth factor" molecule. In cancerous cells, however, the protein that initiates the reaction has been altered and remains active even in the absence of the growth factor. Once they have identified the defective protein, the students can "drop down" to BioLogica™'s molecular level, to examine it and see why it behaves so anomalously. By tracing the flow of the molecule backwards they can observe it being created from DNA, and they can use this information to determine which gene produces it. Finally, by comparing this gene to the normal one, they can locate the mutation and perhaps even identify, from a "rogues gallery" of suspects, the retrovirus that caused it.

Scripting BioLogica™

A piece of software as complex as BioLogica™ can be difficult to use in the classroom. For one thing, each investigation, such as the one we have described, requires a particular configuration of the program - cells must be set up with appropriate pathologies, available instruments must be carefully chosen so as to guide the investigation without constraining it unduly. It will be useful if the software itself can pose the problem to the students, and even more useful if it can monitor their progress, offer suggestions, and periodically ask them questions. Most important of all would be to give curriculum designers the ability to construct a suitable context for student investigations by linking them to real world scientific questions or social or ethical issues.

Scripts will enable us to do many things. We will experiment, for instance, with the possibility of providing a real world context for student investigations by having a scientist (or an actor playing one) describe the research problem she is working on and offer encouragement and advice as students proceed to try to solve it. We will exploit the scripts' ability to monitor student actions by attempting to identify "teachable moments" and providing feedback. In the scenario described above, for instance, we could have the script react when the student team isolates a particular protein or identifies its gene.

An important goal of our project is to simplify the writing of scripts, so that teachers, educational researchers, and curriculum developers, with no expertise in programming, will be able to write their own scripts or modify those of others. We will explore the use of "wizards" that will guide script developers possibly through a dialog in an interview format. Our goal will be to make script writing no harder than absolutely necessary. Though it may be a challenge to produce pedagogically useful scripts, at least we will make it easy to produce scripts that "work" (in the sense of not crashing the computer!) and do what the author intended.

BioLogica™ is being produced by Bob Miner and Ed Burke, using Java on the Windows platform. Eventually, we plan to port the software to the Macintosh platform. Bob is responsible for the underlying engine and associated user interface, and Ed is implementing the EASL environment. They expect to have a demo version, which will largely recapitulate the functionality of GenScope, by July 1998, and a working program by the end of the year. Other members of the BioLogica™ team include Paul Keefe, who writes scripts for GenScope using AppleScript (a time-consuming and frustrating process that confirms the importance of EASL). Joyce Schwartz and Joanna Lu are designing scripts and producing paper-and-pencil prototypes. Mary Ann Christie is studying students using GenScope and will shortly do the same for BioLogica™. I watch over the group and write articles like this one.

Paul Horwitz is Principal Investigator of the BioLogica™ Project, which will be available in Spring 1999. gsinfo@concord.org⁷

Spring 1998 | Table of Contents² | Library Index¹ | CC Home⁴

Copyright © 1998 The Concord Consortium, All rights reserved. Last updated: 12-July-99
Questions and comments regarding this site can be sent to webmaster@concord.org¹¹