Science Bleg

Testosterone from ballsI HAVE CONCEIVED this desire to put an image of a model of a testosterone molecule on the cover of my novel, The High-T Affair. Being an acquisitive sort, I saw a ball-and-stick model of one in the Wikipedia article and decided I had to have one of my own.

Easy peasy, I thought. After all, I have several 3D packages, including the big kahuna, 3DS Max, and Modo 601. I should be able, with very little google.fu, to find a model that I can import into a scene and use there.

Not so much.

OK. There should be a way to find out how to make one from scratch. And, sure enough, there’s all kinds of packages. Problem is, I suspect, they assume more knowledge than I have or could get within a reasonable period of time. I might want to get it later, but right now, I have limited time.

So, before I abandon this quest for good and ever (or do something else), I thought to ask youse guys if anybody might know where I could get one or learn how to make one in a bog standard 3d package without boning for my piled higher and deeper. Anybody? Beuller?

9 responses to “Science Bleg

  1. 3ds is a weak package unless you’re doing organic shapes. Solidworks. Deneb, Catia, Cadra, and Alibre are all engineering packages, and that’s an engineering drawing. I could draw it for you and give it to you in a format you could use, manipulate around, etc. Alibre will still let you draw for free for 30 days, and it’s butt simple to use.

  2. Mark Philip Alger

    Your offer is too kind, but I have a feeling I should make sure — this being a wholly commercial effort — that everything in the cover art is entirely of my own making.

    It might even prove salutary that I figure out how to fake it in modo. I could use the practice. It’s not, after all, like I need to be able to mess with the valance of the bonds or whatever. I can literally just draw a bunch of balls and sticks in the right configuration and stick ’em together.

    I was just hoping that someone such as yourself might know, “Oh, yeah. There’s a thing at Scientific Rugrat for ten bucks — an introductory organic chemistry blah-blah that has a Wavefront Object export filter.” Or something the like.

  3. Alibre design.

    30 days, if you give it a half hour or more a day, will be plenty adequate to do what you want.. I don’t know what they’ll let you export to.

    I know plenty of people who write full time for a living, and not a single one has ever designed the artwork for their book covers. I also know several of the people who do artwork for book covers, and they’re the scum of the earth. I figured I’d fit in fine. I can talk you through doing this in Alibre very quickly.

  4. Mark Philip Alger

    Alibre won’t let me install. It tells me I’ve already had my bite at the apple.

    The authors you know may not do their own covers because they’re not allowed to. But they’re also not allowed to write more than two books a year, they have screwy non-compete clauses in their contracts, they get paid royalties that suck donkey dicks — if they even get an honest accounting from their publishers.

    That’s all changing. I think that, in the future, the author will be the publisher, and the choices of cover artist, book designer, editor, printer, distributor will all be exposed to him, even if he chooses to delegate to someone else.

    I may be nearly unique in that writing, while my primary life goal for 50 years, is not my career. I am a designer. It’s what I do for my living, and I’m damned good at it. I see no reason to abdicate to someone else.

    Acually, what might help most would be some analysis of the model above. What are the relative radii of the spheres and sticks? What is the lateral angle? (Obviously, the carbon ring is a hexagon, so the polygon corners are at 60 (I mean 120) degrees, but what’s the side-slip value?) And so-forther.



  5. I have access to molecular modeling software at my work, but it’s big bucks. I did try out Jmol (, once you have the molecular structure file it looks like it may do what you want. I created a testosterone model here, and I can e-mail or ftp you the molecular structure file if you want to give that a try.

    As you and og are discussing, you can also use any mechanical 3D CAD package to create something equivalent. However, getting the bond angles, van der Waals surfaces and other technical details correct may be a headache. The jmol package looks ‘right’, the flip side is that after getting all the molecular details correct you may end up with a dull graphic.

    Alternate solution: Go to the craft store, get some thin wooden sticks, styrofoam balls and paints. Build a real-world 3D model, then have Dolly hold it just about chest high for a cover shot. No one will notice any imperfections in your testosterone model. (The craft store one, that is.)


  6. Actually, the just-looks-the-part is fine with me.

    If you look at the article on Wikipedia, they also have a flat graphic of the same molecule. I was considering reproducing that in Illustrator format and rendering it as something like gold wire embedded in stone. If I were to get a model built, I’d probably surface it with something like a gold mirror, rather than the low-visibility black-and…

    So what do you think the angle on the hydrogen bonds is? 30 degrees? 60? Do they just alternate? Or hook on at random? Or… what?

    I wonder if Dolly would be willing to have a henna “tattoo” of the wireframe model on her boob? Whattaya guys think?


  7. On bond angles and atomic radii: Now it starts to get a bit more complicated. Most of the carbon atoms in your structure above should have bond angles of ~109.5 degrees in a tetrahedral configuration. Starting one carbon clockwise from the -C=O carbonyl group at the lower left, all the C-C and C-H bonds should be ~109.5 going clockwise up toward the last five-member ring. Your structure correctly shows the ‘puckering’ of the two six-carbon cyclohexane rings, where all the bond angles should be 109.5. Once you get to the five member cyclopentane ring, bond angles become more distorted as the tetrahedral carbon bonds are strained. One purpose of some molecular modeling software is to try and predict the bond angles and lengths which results in a potential energy minimum.

    In the first cyclohexane ring at the lower left, there are three carbon atoms which have double bonds. In this bonding mode, the bonds are planer at 120 degrees. Again, strain from nearby groups and distorted bonds may change this slightly, but the double bond can not be ‘stressed’ as much as the single-bonded carbon before it falls apart.

    As to the atomic radii, in spite of what all these models appear to show there are no ‘hard’ atomic radii, only an increasingly diffuse electron cloud as you move away from the atom. Synthetic organic chemists often like to work with the stick or ball-and-stick models, as it allows them to see the bonds they are trying to manipulate in their chemical reactions. Biochemists, on the other hand, often work with ‘spacefill’ models, which hide the bonds but give a better view of how this fuzzy electron cloud interacts with other macromolecules, such as enzymes and proteins. A good chunk of biochemistry research is studying how this ‘lock-and-key’ mechanism works and how it can be manipulated. In short, there is no fixed value; the best case I would suggest is doing what you think looks good.

    I don’t think Jmol or my software at work will do metallic gold surfaces, but I’ll see what I can do and try to get you some assorted jpgs. Perhaps you could use them as a starting point.

    As you can see, you can get yourself tied up trying to get things ‘scientifically accurate’ and lose the objective of the cover graphic: to grab the potential buyer’s attention, curiosity and desire. I’m not big into tattoos on women, but I’d be willing to make an exception for a henna tattoo on Dolly.

  8. Actaully, Jay, you’ve given me enough clueage to figure out the rest. (But, wow! I’m amazed I remember so much of my HS organic chem after 40+ years.)

    Not so much worried about being technically accurate as I am wanting to get as close to this as possible with the least effort. But I’m figuring I’ll model these as a collection of spheres and cylinders.

    Dolly wore mehndi (henna tattoos) in the Between the Lines pastiche (ref: X:WP 4th Season ep), in which she and Drummond finally meet as both of them fully adult humans. Drummond abhors in-skin ink and piercings and asserts if Dolly ever gets either, he’ll leave her.


  9. Clean out your system registry. Then Alibre will let you install.