<DIV>Check out 2D DIGE technology from Amersham!<BR><BR><B><I>MyungHo Kim <firstname.lastname@example.org></I></B> wrote:
<BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">2D Electrophoresis Gel Image and Diagnosis of a Disease<BR><BR>The process of diagnosing a disease from the 2D gel electrophoresis<BR>image is a challenging problem. This is due to technical difficulties of<BR>generating reproducible images with a normalized form and the effect of<BR>negative stain. Here, we discuss a new concept of interpreting the 2D images<BR>and overcoming the aforementioned technical difficulties. This concept makes<BR>use of 2D gel images of proteins in serums and we explain a way of<BR>representing the images as vectors in order to apply machine-learning<BR>methods, such as the support vector machine, a decision tree, and neural<BR>network etc. (For details, see www.biofront.biz or<BR>http://arxiv.org/abs/cs.CC/0305048)<BR><BR>1. Representation<BR><BR>I) Taking the whole image<BR><BR>By enumerating the whole set of numbers (densities) corresponding to each<BR>pixel
in a predetermined order, we will represent an image as a vector in a<BR>finite dimensional Euclidean space.<BR><BR>II) Choosing spots<BR><BR>Choose a finite number, K, of conspicuous spots representing proteins and<BR>their quantities, for example, we may take CA-1, BD-1 CA-2 and CA-3. Each of<BR>these chosen spots will have a corresponding number, which is the sum of the<BR>numbers assigned to each pixel consisting of the spot. Thus, the sum of each<BR>spot will represent the relative quantity of the protein corresponding to<BR>the spot relative to other spots. By enumerating the quantities of those<BR>four proteins, we have a vector in the four dimensional Euclidean space.<BR><BR>Discussions: At a glance, in representing an electrophoresis image, the<BR>second method seems more natural than the first one. However, though<BR>considering the quantities of proteins looks intuitive and appealing to<BR>biological meaning, the procedure of measuring the relative quantities of<B
R>chosen proteins may not be accurate for our purpose. On the contrary,<BR>accepting the whole image could contain more than we realize. We all know<BR>from a meticulous analysis that recognition of a person with a picture is<BR>due to the human brain¡¯s ability of computing relative positions of<BR>specific objects such as nose, eye, mouth, ears, distance between eyes. Each<BR>pixel with its own density plays a role as a member of a whole image. Though<BR>each pixel does not give any clue by itself, all pixels together with others<BR>send us a concrete picture we conceive. Therefore, it is reasonable to say<BR>that the intrinsic invariants of an image are the relative position of a<BR>pixel with its density. The first method is about considering the whole<BR>package of all relative positions and their densities.<BR><BR><BR>2. Problems in 2D gel image and its staining methods<BR><BR>2D gel electrophoresis is a method that separates proteins in a<BR>2-dimensional plane by mas
s and pH of proteins. As is often the case in the<BR>most of experiments, there are two technical problems we have to compromise<BR>so that the process of numericalization is acceptable and tolerable.<BR><BR>1. When the amount of a certain protein reaches a ¡°threshold¡±, the<BR>silver stain density decreases. This phenomenon is known as the negative<BR>staining effect.<BR>2. Even if the experiments are performed carefully, there always will be<BR>some variations of images. For example, in the image, the same protein will<BR>not be in the same position relative to other proteins.<BR><BR>Discussions: Conceptually we need to find out a method of representation<BR>of each serum, which is reproducible with some tolerable variations. This<BR>attempt, as we did in the previous discussion, is feasible. Whatever<BR>qualities we observe, estimate or sample, there are variations in<BR>measurements or recording, since everything changes and it is impossible to<BR>produce the equivalent
results every time. For example, our heights changes<BR>at morning and night, and blood pressures vary every hour. Likewise, there<BR>are variations in 2D gel images even under the assumption that all the<BR>experiments are perfectly accurate and executed the exactly same way. It<BR>could be caused by the status of a donor of serums or experimental setting.<BR>However, patterns of the variations seem quite consistent and, if we ponder<BR>our heights in micrometer, then it is even natural to accept an image as it<BR>is.<BR><BR><BR><BR>_______________________________________________<BR>BiO_Bulletin_Board maillist - BiO_Bulletin_Board@bioinformatics.org<BR>https://bioinformatics.org/mailman/listinfo/bio_bulletin_board</BLOCKQUOTE></DIV><p><hr SIZE=1>
Do you Yahoo!?<br>
Free <a href="http://us.rd.yahoo.com/mail_us/tag/*http://calendar.yahoo.com">online calendar</a> with sync to Outlook(TM).