Part Type:Coding:Icon Discussion

possible enzyme icon? Part icon enzyme1.png...basic idea is a box binding two substrates in it...(--mel)

  • I like the little shapes in the box, but it would be better to omit the top of the box (make it a trough instead). That's more realistic for most enzymes. Also, could the box sides be extended to convey the notion that the enzyme is usually much bigger than the substrate(s)? [SCM - 3/12/07]
  • OK, I made a few "open trough" format enzymes....closer? I'm still not sure what the metabolism ones should be like Enzyme open.pngI rounded the bottom to align better with the stuff inside it
    • This one looks very nice. I like the shape complementarity -- perhaps you can work with it a little more (use a triangle rather than a ball for one of the substrates? Also, rounding the ends of the open box would be good (proteins are rounded and irregular -- with no square corners. [SCM 3/28/07]
  • Another image, this one more symbolic with the protein coding parts, but the coding regions aren't necessarily substrates, so it could be misleading Enzyme open2.png
    • I agree about the possible confusion of using the arrow (which we should restrict to icons for DNA.) Let's discard this one. [SCM - 3/28/07]
  • An alternative suggestion from Reshma: Make little curved arrows and put several of them in a row, all pointing in the same direction (reminiscent of a metabolic pathway). [SCM - 3/14/07]
  • Metabolism
  • would metabolism really represent the action of an enzyme? I don't think I've ever seen the curved arrows before...are those the ones indicating electron-arrow pushing? That idea with the trough sounds good to me. --Smelissali 15:29, 26 March 2007 (EDT)
    • Well, enzyme catalysis always causes a chemical change and "metabolism" constitutes the sum total of all (biologically programmed) chemical changes in the cell/organism. But I agree about the lack of familiarity of most people with presentations that have curved arrows (they are generally used in diagrams where the process is shown vertically, not horizontally). [SCM 3/28/07]
  • Possible repressor protein icon? Part icon inhibitor2.png, Part icon inhibitor3.png?
  • Repressor icon should be an inverted T, as in circuit diagrams. Activator icon should be an arrow pointing up. [SCM - 3/12/07]
  • Here's a simple repressor bar, is this what you were thinking of? Repressor bar.png --Smelissali 08:41, 27 March 2007 (EDT)
    • Yes, that's a good step in the right direction. Could you put the little "T" upside down on **top** of the bar? Also, if you can add a short horizontal handle, that would be perfect. The activator icon could just be a little arrow (again with a horizontal handle).
  • Reshma and Mack favor a slightly more elaborate symbol for each of these, along the lines of the circuit-component interaction lines. each would have a short vertical bar, a horizontal bar starting at the top end of that, then from the end of the horizontal bar a second, equal-sized vertical bar going downward. For activation the tip of the second bar would be an arrowhead, and for repression it would be another horizontal bar. [SCM - 3/14/07]
  • Is there a special reason for having two substrates 'bound' to the enzyme box? Some (if not most) enzymes would only bind one substrate at a time... Marko August 6, 2007

I would like to open a wider discussion of icons than presented above only for protein types. In part this refers to the recent discussion on the Help_talk:Regulatory page but wish to make it very general - hopefully not too general for the current moment in the 'evolution' of the Registry. SB has the power of destructuring, assembling and restructuring and the concept of Registry is to keep information and material of genetic parts and devices. Each engineering genetic element was appointed an icon that simplified understanding of complex devices. Parts were ordered by categories which were reasonably well described. In this context, my points are as follows:

  • There is currently no clear hierarchy of icons (they are all of about the same size and complexity and some even appear in different categories). We should establish icons that would clearly differentiate between Parts, Devices and Systems.
  • Icons are needed for separate regions of Parts. Although some parts clearly serve only one function (e.g. ribosome binding sites), some Parts are in fact Devices - an evident example is the Regulatory part ('promoter'), which currently includes both promoter and operator. One single icon cannot discriminate between repressor and activator binding regions attached to the promotor sequence. A discussion could point out those subparts (part regions or subcategories) that would need separate icons.
  • Iconization of subparts would help understanding complex parts and would promote SB concept of genetic elements beyond SB community. There is currently no standardized way of depicting individual genetic elements in the life science community but it would be helpful to have some standards here as well (current icons developed for the Registry can be the starting point). Once we agree upon icons, we should have them available as high(er) resolution elements that could easily be included into presentations and articles.
  • Registry does not need to be restructured so as to include part subregions as separate BioBricks. Icons could remain abstraction tools and there could be far more icons than part categories.

Marko August 6, 2007

Marko's points are well-taken. We definitely need an upgrade for Registry icons. Are there any gifted artists out there who have the skills to draw some candidates to replace (and extend) the current set? Perhaps vectored drawings would be better than rasterized ones... As for the other issue (icons for subregions), I think we need to consider this very carefully so as not to confuse people about what can be assembled and what cannot. Perhaps the style (size, textured fill, etc.) of icons that correspond to true BioBricks can be clearly distinguished from some (less prominent) subpart icons for operators, activator binding sites, etc. Note also that this issue merges with the question of an ontology for synthetic biology (a search for "ontology" on OpenWetWare yields a large number of informative hits). We need more discussion and an SB ontology 1.0... Scott August 11, 2007.

While I agree with much of what is being said above, the existing part icons are not that bad! As Tom Knight has often pointed out, we need multiple different representations for devices/systems. The DNA sequence view currently used on the registry seems like a reasonable way to represent the assembled parts. Functional representations of devices/systems migh require more complex sets of icons but I think the existing promoter/RBS/coding region/terminator icons are very useful. it would be great if there were slightly higher resolution versions available! -Bcanton