|
12/18/2003
Complex Maps and Curveballs
This is my last column for 2003. Brian Trumbore is giving old
Bortrum a two-week Christmas-New Year vacation. As I start
this column on Sunday, December 14, pictures of a tired and
disheveled Saddam Hussein are flooding the media. This story
certainly dwarfs what was the biggest story in our New York
metropolitan area, at least among sports fans, the loss of Andy
Pettit to the Houston Astros. Yankee fans will miss those intense
eyes peering from behind his glove as he reads the catcher’s sign
for a sinking curveball. Now those fans also have to contend
with a rumor that Roger Clemens is considering coming out of
retirement to join his former teammate on the Astros.
No doubt, most of these fans missed an article in the November
issue of the American Journal of Physics that I found referenced
in a short item by Adrian Cho in the December 5 Science. Mont
Hubbard of the University of California and his colleagues have
challenged one of baseball’s long held beliefs, namely, that a
fastball will travel farther than any other pitch when struck
properly by a competent batter.
Hubbard et al. calculate that a curveball “smartly struck” is more
likely to end up in the bleacher seats than a fastball similarly
struck. They reason that a ball with backspin travels farther
because of the uplift due that backspin. We golfers can accept
that spin plays a very important role in determining the path of a
ball; witness my repeated duck hooks into the water on those
Florida courses. The researchers claim that, since a curveball
already has the appropriate backspin, it’s going to outdistance the
fastball by about 12 feet. There already seems to be some debate
about whether the curve has to be a “hanging” curve up around
the letters or whether a ball breaking down and away can be
stroked just as far. Gary Matthews, who hit 234 homers in his
big league career, is quoted as saying that catching such a pitch
just right can cause the ball to “go a long way.”
I’m not qualified to weigh in on this issue. During my own
baseball/softball career, I hit only one homerun over the wall.
Actually, it was over the logs marking the boundary of our
softball field at NACA, Lewis Flight Propulsion Lab in
Cleveland. I’m even less qualified to comment on the content of
another article in the same issue of Science that really blew my
mind. Scientifically, in my opinion, the year 2003 was
highlighted by two stories - the completion of the mapping of the
human genome and by the determination of the age and
composition of our universe. The latter achievement confirmed
conclusively that our universe is overwhelmingly dark matter
and dark energy. However, a huge question remains. What is all
that dark stuff?
Similarly, the completion of the human genome map delivered
the composition of our DNA but opened up some very difficult
questions. How many genes are there? What are the functions
of all the “junk” in our DNA between the genes? (Is this junk
the “dark matter” of our DNA?) Most challenging, what are the
interactions among the thousands upon thousands of proteins
churned out in our cells based on our DNA’s instructions? These
interactions determine our life and our health and understanding
them is necessary for developing new drugs and treatments for
the diseases that plague us.
When I opened my copy of Science, I turned just a few pages
when I ran across a foldout section and I shuddered. Previous
foldouts in Science have included the aforementioned human
genome map, which is now covering half a wall in our quarters at
UMDNJ Robert Wood Johnson Medical School. None of us, or
any of our visitors, has had any reaction to the map except to
stare at it in awe and walk away.
With trepidation, I opened this foldout and I was totally
overwhelmed. The foldout was supplied by the CuraGen
Corporation, a genomics-based drug development company and
was titled as “A Protein Interaction Map of Drosophila
melanogaster”. You may recognize Drosophila m. as the
common fruit fly, perhaps the most studied insect in science.
The associated article of the same title had 49 authors by my
count, most from CuraGen, others from the Yale and Wayne
State Universities’ Schools of Medicine. A subtitle on the
foldout is “The Dawning of the Age of Systems Biology
Medicine in Context”.
To explain why I was overwhelmed by the work, let me try to
describe this protein interaction map. Any frequent flier has seen
in the airline magazines the maps of the routes the airlines fly.
These maps have dots or circles representing various cities with
lines connecting the cities served by the airline. The hub cities
have many lines radiating from them. For a large airline serving
many destinations, the map can be quite “busy” in appearance.
Now imagine that an airline has serves 2,346 cities joined by
2,268 routes. Whoa! “Busy” would be putting it mildly!
Our foldout replaces cities with 2,346 proteins, each protein
designated by a circle filled in with one of 14 different colors.
The color represents a protein that is localized in particular areas
or structures of the fruit fly’s cells. For example, green circles
represent proteins found in the nuclei of the cells. Yellow circles
are proteins associated with mitochondria, the powerhouses of
our cells. Some proteins are associated with both nucleus and
mitochondria and hence are half yellow-half green. Each
circle/protein has a name, for example, “put”, “rig2”, “CG4860”,
“Mnt”, etc.
These protein circles are joined with lines of different colors that
indicate interactions among the proteins. (I won’t complicate
things by trying to explain the different colored lines.) As with
our airlines map, certain proteins are “hubs” that interact with
several proteins, some hub proteins having interactions with
perhaps 10-15 other proteins. Reading the paper, I was surprised
to find that the foldout map was a “filtered” map that eliminated
interactions that didn’t meet certain requirements. Another map
in the paper seemed to me of more direct value to us humans.
That map, which includes 3,000 interactions among 3,522
proteins, highlights fruit fly proteins that are similar to proteins
associated with human diseases.
I have barely touched upon the complexity of these protein
interaction maps or the “yeast two-hybrid-based” approach used
to arrive at the maps. Suffice to say that the authors use the maps
to draw parallels between protein interactions in the fruit fly that
have their counterparts in humans. Interactions mentioned in the
article include sex-determination, insulin signaling, calcium
regulation, eye development and innate immunity. We have a lot
in common with this insect.
The complexity of this protein interaction map for the lowly fruit
fly illustrates the daunting challenge to those who hope to map
the human proteome, the interaction map for all the many
thousands of proteins programmed in our DNA. You can bet
that 2004 and the years following will see a host of papers on
protein interactions in all kinds of life forms. Let’s hope that
these esoteric papers are balanced by the occasional more down
to earth works dealing with such subjects as a bat hitting a
hanging curveball.
Have a great holiday. Bortrum will be back on January 8, 2004.
Allen F. Bortrum
|
