Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Biotech / Medical : AFFYMETRIX (AFFX) -- Ignore unavailable to you. Want to Upgrade?

To: jerryriti who wrote (740)	6/30/1998 12:15:00 AM
From: Richard Haugland	Read Replies (4) \| Respond to of 1728

Jerryriti: Sorry that patent claims have to be written pretty much this way. For instance one cannot use a period in a single patent claim until the very end, even if the claim is a page or more long. Lets see what I can do: The claim stated: An array of oligonucleotides, the array comprising: a planar, non-porous solid support having at least a first surface; and a plurality of different oligonucleotides attached to the first surface of the solid support at a density exceeding 400 different oligonucleotides/[square cm], wherein each of the different oligonucleotides is attached to the surface of the solid support in a different predefined region, has a determinable sequence, and is at least 4 nucleotides in length. The claim covers a checkerboard wherein there are exactly four colors of pieces (A, T, G and C for the 4 nucleic acid bases adenosine, thymidine, guanosine and cytidine) except that they are stacked vertically on each and every square. No two squares have exactly the same sequence of chips stacked on them but the exact sequence on each square is known and recorded in some file. The AFFX patent covers "checkerboards" where there are at least 400 squares per square centimeter, which means that each square is about 1/2 millimeter by 1/2 millimeter or likely even smaller. Like computer chips, small is usually beautiful in the screening business. The "...at least 4 nucleotides in length" refers to the number of "checkers" that are "stacked" on each square. When I referred to "covalent" it means that the first "checker" (i.e. nucleotide or "nucleic acid base", which means pretty much the same thing) is "stiched down" to the board by a chemical bond. Furthermore the checkers are bonded to each other as they are built up. Thus one nucleotide becomes a polymeric oligonucleotide. AFFX controls the sequence in which each square is stacked using a light-directed synthetic process and thus always knows which bases are stacked on which square. For instance a stack 5 bases high on the first square could be A-A-A-A-A and the second square may be A-A-A-A-T, the third A-A-A-T-A, the fourth A-A-T-A-A etc in one direction and perhaps the last square is C-C-C-C-C. In between there are a lot of mixed combinations such as C-T-G-G-A. Of course the numbers of unique combinations is large for even a 5-mer. This is what AFFX sells as a chip. When used to detect a genetic sequence the square with the A-A-A-A-T sequence for instance will bind targets for DNA in the test sample that have the sequence T-T-T-T-A (the complementary sequence) if they are present but will not bind sequences such as G-G-G-G-C. The square with the C-T-G-G-A will bind target with the sequence G-A-C-C-T. Presence of the matched pair is usually detected by attachment of a fluorescent dye or by other means. What the patent means overall is that any chip that has squares of size less than 1/2 mm by 1/2 mm (i.e. moderately high density) and "stacks" of more than four bases is covered by this patent if it is tied down to the surface by a covalent bond (if it is on an impermeant surface). Some companies make their "stacks" (or oligonucleotides) off of the chip then deposit them on the surface from a solution (such as by using an ink-jet printer!) The post from Perkin Elmer and Hyseq today seems to be done that way. I am not certain how they keep these from washing away when they mix them with the target sample. It seems to me if they attach them down by any means that it would violate this patent (if their density is >400 squares per square cm). Four or five bases is about the minimum cutoff for utility for forming useful complementary pairs in actual applications. The 1989 date on the original patent application, of which this is a continuation in part, means that for another company to claim that they owed this technology they would have to have invented it before that date, which was very early in the game of chips. The AFFX patent seems broad and important to me. It is like having a patent on all computer chips that will ever be more dense than this "checkerboard." Of course there can be future technology improvements that would, for instance, allow one to increase the density to perhaps 100,000 squares per square cm that could be patented by another company but because that would read on the AFFX patent of greater than 400 squares per square cm, the other party could not practice their invention without a license from AFFX. There is a somewhat practical limit on the smallest size of any "square" in that the detectable signal that occurs when the target binds to the chip (the "hybridization step") is pretty much proportional to the area of the square. I don't, however, know what this practical limit is and it will depend to some degree on the brightness of the dye used for the detection step. Incidentally the word "comprising" in the first line is a useful patent term (if one is the inventor) because it means, in general, "...including what I say is needed and anything else, whether I know about it or not..." Hope that clarifies it somewhat.