[00:00:04] Speaker 01:
Our last case for argument this morning is Quake versus Lowe, case number 18-1779.

[00:00:10] Speaker 01:
Mr. Reines, you reserve three minutes of your time for rebuttal, correct?

[00:00:18] Speaker 04:
Yes, thank you, Your Honor.

[00:00:20] Speaker 01:
You may proceed.

[00:00:21] Speaker 04:
Thank you, Your Honor.

[00:00:22] Speaker 04:
Ed Reines on behalf of Professor Stephen Quake and Christina Phan.

[00:00:27] Speaker 04:
Although the board correctly followed this court's instruction and guidance and came to the conclusion that random DNA sequencing was a molecular counting method, even a chromosome counting method, disclosed in the patent in columns 19 and 20, it legally erred by fundamentally getting the law wrong.

[00:00:52] Speaker 04:
How did that happen?

[00:00:54] Speaker 04:
The court decision turned on its conclusion that this case was like Santa Cora and Lily, where there was just a wish, that it was a wish for Quake and Fan to think that you could use random DNA sequencing as the chromosomal counting method.

[00:01:11] Speaker 04:
for the invention.

[00:01:13] Speaker 04:
And that's completely lacking in the record.

[00:01:16] Speaker 04:
The technology was disclosed in 2003.

[00:01:19] Speaker 04:
There were products in the market that's all in the patent application.

[00:01:23] Speaker 04:
It's in the definition agreed upon definition of skill in the art was that person's skill in the art and their basic competency with no random sequencing.

[00:01:31] Speaker 04:
So the idea that this was a new technology is just a factually false statement.

[00:01:36] Speaker 04:
They don't even cite anything.

[00:01:37] Speaker 04:
The board doesn't.

[00:01:38] Speaker 04:
Um, so that's, it just goes way off.

[00:01:41] Speaker 04:
And then what they do that is they use that analogy to send a corny, by the way, Santa corn, Lily, we're talking about edge biotech cases where they don't even know if the molecule is going to exist.

[00:01:52] Speaker 04:
It's totally speculative.

[00:01:54] Speaker 04:
The idea that when Quake says, here, I'm going to show you how my invention works.

[00:01:57] Speaker 04:
This is a brilliant invention of rather than trying to separate maternal DNA and fetal DNA, you put them together for 10 years.

[00:02:06] Speaker 04:
low and everyone else was trying to separate them.

[00:02:09] Speaker 01:
So, Mr. Reinus, when we look at claim one of the 18 pattern, and we look at paragraphs A, B, C, and D, the way I understand that the, in my view anyway, one of the primary issues here in your case is that there's something missing between C and D, and that would be something, the normalization of data.

[00:02:35] Speaker 01:
There's something missing.

[00:02:37] Speaker 01:
It says using the data of step C. And looking at step C, it's just identification of the chromosomes.

[00:02:47] Speaker 01:
There's nothing else there.

[00:02:49] Speaker 04:
Your Honor, I look at it differently.

[00:02:51] Speaker 04:
I think it's really, in D is you do what's obvious.

[00:02:54] Speaker 04:
So you have ways to count chromosomes and in any method you have to normalize because you're going to get more hits from chromosome 21 if it's big than chromosome 20.

[00:03:03] Speaker 01:
So let's say that, so your argument is that since formalization was well known and

[00:03:09] Speaker 01:
Understood.

[00:03:10] Speaker 01:
Yeah, undisputedly.

[00:03:11] Speaker 01:
Yeah, that a post-ita would probably know that.

[00:03:13] Speaker 01:
That still doesn't excuse a total absence or lack of disclosure of normalization in the pen, does it?

[00:03:22] Speaker 04:
It totally does, because normalization's not required.

[00:03:25] Speaker 01:
How can it?

[00:03:26] Speaker 01:
That means that we just, instead of reviewing the pen, let's just see what the post-ita knows.

[00:03:31] Speaker 01:
I mean, the pen has a purpose for disclosure.

[00:03:35] Speaker 01:
There's a purpose for written description.

[00:03:37] Speaker 01:
How does a written description here assist you with respect to normalization of data?

[00:03:43] Speaker 04:
Good question.

[00:03:44] Speaker 04:
The answer is, in looking at a written description issue, you have to look at the nature of the invention and the predictability of the R.

[00:03:54] Speaker 04:
And so the idea is, is it a speculative thing?

[00:03:57] Speaker 04:
Is it just an inventor saying, well, let's create fusion or something like that?

[00:04:02] Speaker 04:
Or do they possess the invention?

[00:04:04] Speaker 04:
Here, the invention, and this is what I was getting to a moment ago, was you don't separate the maternal and the fetal DNA.

[00:04:10] Speaker 04:
No one ever did that.

[00:04:11] Speaker 04:
Everyone was trying to divide it.

[00:04:13] Speaker 04:
There's nothing in this record that anyone ever did that before Professor Quake's major invention.

[00:04:18] Speaker 04:
That's an issue here.

[00:04:19] Speaker 04:
No one ever did that.

[00:04:20] Speaker 04:
He said, I'm showing you how you do it with digital PCR.

[00:04:24] Speaker 04:
Admitted.

[00:04:24] Speaker 04:
He laid it out chapter and verse.

[00:04:26] Speaker 04:
He spent a lot of time.

[00:04:27] Speaker 04:
And then he said, and this is, I think, as shown by the table of contents for the application, there's a variety of ways where you can do the chromosome counting.

[00:04:36] Speaker 04:
You can do chromosome counting this way.

[00:04:37] Speaker 04:
You can do chromosome counting that way.

[00:04:39] Speaker 04:
You can use DNA sequencing.

[00:04:40] Speaker 04:
Two columns worth reference to commercial products.

[00:04:44] Speaker 04:
Now, the point being that if it's mundane technology,

[00:04:50] Speaker 04:
That's within the basic competency, like you count chromosomes by reference to the product and the patent application from five, seven years earlier, whatever it is, the 2003 application.

[00:05:02] Speaker 04:
No one's saying that there's anything inventive about how you do step D. The question is, step D requires you take the number of hits you got for chromosome 21,

[00:05:13] Speaker 04:
the suspected aneuploidy, and the number of hits you got for chromosome 20, and you compare them.

[00:05:18] Speaker 04:
If you've got more 21, uh-oh, you may have Down syndrome, baby.

[00:05:20] Speaker 01:
So step D is a comparison step.

[00:05:23] Speaker 04:
Between the suspected aneuploidy and not.

[00:05:26] Speaker 01:
Step C is identification step.

[00:05:28] Speaker 01:
What's missing here, and this is what I'm trying to get you to focus on, is the normalization of data step, and with respect to massively parallel sequencing.

[00:05:41] Speaker 04:
But that's an implementation detail, which is the other side calls it a detail in their appellate brief.

[00:05:49] Speaker 04:
They call it a detail normalization, your honor.

[00:05:52] Speaker 04:
it's part of step D's statistical of how you do it.

[00:05:56] Speaker 02:
It's not a separate step.

[00:05:58] Speaker 02:
My question is, I think, similar, but I mean, phrased a little bit differently.

[00:06:03] Speaker 02:
If everything you're talking about in this specification is largely devoted to targeted sequencing, you may dispute that, but I just take that as... But I think that was the last federal circuit decision.

[00:06:17] Speaker 02:
Well, it is, but it isn't, because the problem is

[00:06:21] Speaker 02:
You've shown all these steps to do targeted sequencing and I may be getting the science wrong.

[00:06:30] Speaker 02:
It's digital PCR.

[00:06:31] Speaker 01:
Let me finish.

[00:06:33] Speaker 02:
If you don't have to do this kind of normalization when you have targeted sequencing, there's no reason for the specification

[00:06:42] Speaker 02:
to claim it.

[00:06:43] Speaker 02:
I get that the normalization process we're talking about is well known in the art, but if it's not in the specification, why doesn't that still lead to the conclusion that the only thing the written specification supports overall is targeted sequencing?

[00:07:01] Speaker 04:
Because it's an implementation detail of how you do the comparison.

[00:07:04] Speaker 04:
You have to do statistical analysis, whether you're using targeted sequencing, which isn't disclosed in this patent application, or digital PCR, which is the main embodiment.

[00:07:13] Speaker 04:
I think that's what your honor was referring to.

[00:07:15] Speaker 04:
In all of those, you have to do all kinds of statistical analysis.

[00:07:17] Speaker 04:
It's in the patent, the T-test, the student T-test.

[00:07:20] Speaker 04:
Everyone has to do that.

[00:07:22] Speaker 04:
The point is, would a person of skill in the art looking at this think that this patent application discloses the use of random sequencing as the chromosome counting method for the invention?

[00:07:35] Speaker 04:
And there's no way in the world they wouldn't.

[00:07:37] Speaker 04:
It says,

[00:07:38] Speaker 04:
One way to use the invention is to use random DNA sequencing.

[00:07:42] Speaker 04:
So they clearly would know that that's being attempted.

[00:07:44] Speaker 03:
Where does it say one way to do the detection of aneuploidy is through random DNA sequencing?

[00:07:52] Speaker 03:
I think that's the rub.

[00:07:54] Speaker 03:
If you said that, then we wouldn't be kind of on our hands and knees trying to hunt for clues to find and locate this second method that you're now claiming.

[00:08:05] Speaker 04:
Right, let's see if I can give you a way to do that, which is at 19, column 19, at the part that was quoted by this court in its last decision, at line 59, it states, a methodology useful in the pleasant invention platform is based on massively parallel sequencing random.

[00:08:26] Speaker 04:
Then it goes on, and then it sends you to the Illumina, and then it sends you to the ballast of Bermanian.

[00:08:31] Speaker 04:
And it says, one way to do the invention

[00:08:34] Speaker 04:
The methodology is to use random sequencing, and then continuing to the next paragraph, it says, and then the way you count the chromosomes is you use 30 BP of random sequence information.

[00:08:46] Speaker 04:
Where are you, Mr. Ryneson?

[00:08:47] Speaker 04:
What's that?

[00:08:48] Speaker 01:
Where are you?

[00:08:49] Speaker 04:
The next paragraph.

[00:08:50] Speaker 04:
Well, this is at 20 at line 7.

[00:08:52] Speaker 04:
20 at line 7.

[00:09:04] Speaker 04:
That's the second.

[00:09:05] Speaker 04:
So it says, for the invention, I mean, look, we're going to lose the appeal if you don't accept that it's saying you can use DNA sequencing as the chromosomal counting method.

[00:09:14] Speaker 04:
That's what columns 19 and 20 say.

[00:09:16] Speaker 04:
That's what was decided in the last appeal.

[00:09:18] Speaker 04:
That's what even the board had to agree to.

[00:09:21] Speaker 04:
The issue is, if you have counted the chromosomes, would you know to compare the one you suspect is aneuploid with a control or not?

[00:09:29] Speaker 04:
Yes, you would.

[00:09:30] Speaker 04:
That's why they don't argue it, and that's why the board didn't say, it doesn't disclose the general comparison.

[00:09:36] Speaker 04:
The general comparison is step D. Now, there is no statement of that specific as an example.

[00:09:43] Speaker 04:
So that's what I'll agree.

[00:09:44] Speaker 04:
There is not an embodiment where they lay out, these are the statistical analyses you would do to compare the suspect unemployment to the control.

[00:09:52] Speaker 04:
That doesn't exist in here.

[00:09:54] Speaker 04:
But it does disclose the use of the comparison of the control versus the unemployed.

[00:09:59] Speaker 04:
It's not that complicated.

[00:10:01] Speaker 04:
And then how you do that, people would know.

[00:10:03] Speaker 04:
So people wouldn't say, well, they don't own it because they didn't tell us how you do that comparison.

[00:10:07] Speaker 04:
When it's part of the definition of skill in the art that

[00:10:10] Speaker 04:
that this kind of sequencing would be natural.

[00:10:12] Speaker 04:
They call it a detail.

[00:10:13] Speaker 04:
They don't even want to defend, and they'll say something, but the fact that a person skilled in the art would know.

[00:10:18] Speaker 04:
Their experts said point blank.

[00:10:19] Speaker 04:
A person skilled in the art would know when you're comparing one chromosome to another.

[00:10:23] Speaker 01:
I would agree.

[00:10:24] Speaker 01:
We look at the procedure to...

[00:10:27] Speaker 01:
to be able to express what's well known in the art.

[00:10:31] Speaker 01:
And perhaps in this case, that could be this translation of this data.

[00:10:37] Speaker 01:
But I also believe that there's got to be something in the patent that points to that, something by which we can look at, or someone skilled in the art would look at that and say, OK, here's what you do with the data after the

[00:10:50] Speaker 01:
after the other steps are taken care of, all the way up through C. Here's what you do with the data.

[00:10:57] Speaker 01:
Just something.

[00:10:58] Speaker 04:
You haven't yet pointed out to that one thing.

[00:11:01] Speaker 04:
Right, there is no embodiment of step D that's spelled out.

[00:11:05] Speaker 01:
Or language, or something.

[00:11:06] Speaker 04:
I don't know what language.

[00:11:07] Speaker 04:
There's a language that's... It's undisputed that one skilled in the art would know you compare the aneuploidy chromosome count with the control, because you want to see if it's elevated.

[00:11:18] Speaker 04:
And the implementation of how you do that for this particular embodiment is not spelled out chapter and verse, but everyone would know because it's a detail.

[00:11:26] Speaker 01:
So let's go back to paragraph D. Using the data of step C, and I say, well, what data?

[00:11:34] Speaker 01:
Just the identification of the chromosomes?

[00:11:36] Speaker 01:
That's it?

[00:11:36] Speaker 01:
The chromosomes count.

[00:11:37] Speaker 01:
We know that's not the case.

[00:11:39] Speaker 01:
There's something else that's happening there.

[00:11:40] Speaker 04:
No, it's the chromosome count, the count of the chromosomes, which is from 19 and 20.

[00:11:45] Speaker 04:
You count up how many hits you're getting for 21, you count up how many hits you're getting for 20.

[00:11:49] Speaker 01:
It says, identifying chromosomes to which a sequence has obtained a step B belong.

[00:11:54] Speaker 04:
Yeah, that's the 30 bit.

[00:11:56] Speaker 01:
Okay, then it goes on D, using the data of step C. Right.

[00:12:01] Speaker 01:
Well, there's something that's missing there, isn't there?

[00:12:04] Speaker 01:
No, the input is... The quantification of the data is not missing there?

[00:12:08] Speaker 04:
Right, because you're quantifying it based on the 30 in column 20 where you're going, you're taking your sequences and you're comparing them to the genome and you're figuring out where they're located from and you're counting.

[00:12:19] Speaker 04:
Yes, the count is there.

[00:12:21] Speaker 04:
The issue here is that examples are not required.

[00:12:26] Speaker 04:
That's an Ariad.

[00:12:27] Speaker 04:
You don't need to spell it out if people would know.

[00:12:29] Speaker 04:
The point of the whole predictability, all the considerations that you're supposed to take into account for a written description from Ariad, this is really the point.

[00:12:37] Speaker 04:
This is the why we win.

[00:12:38] Speaker 04:
All those considerations, none of them were made.

[00:12:40] Speaker 04:
They said, this is totally new.

[00:12:43] Speaker 04:
without any support.

[00:12:44] Speaker 04:
If it was new and unpredictable and you didn't know what you were doing, yes.

[00:12:47] Speaker 04:
But you need to take into account predictability.

[00:12:50] Speaker 04:
You need to take into account what the actual invention is.

[00:12:53] Speaker 04:
None of that analysis was done.

[00:12:55] Speaker 04:
Just because an implementation detail on enablement, Your Honor, you keep thinking enablement.

[00:13:00] Speaker 04:
Just because it doesn't have an enabling step doesn't mean it's not written described.

[00:13:05] Speaker 04:
And the fact that it's so trivial tells you that's not a written description violation.

[00:13:10] Speaker 04:
Given the time, I'd like to, I guess, reserve, unless there's a question, which I'd be happy to... Okay, thank you.

[00:13:16] Speaker 04:
Thank you.

[00:13:24] Speaker 00:
May it please the court, the board did not hang its opinion really on Centicorp.

[00:13:33] Speaker 00:
It hung its opinion on a factual finding that this patent specification did not describe this claimed method.

[00:13:41] Speaker 00:
as a unified whole and they said that at appendix page 15 and that fact finding is supported by substantial evidence.

[00:13:48] Speaker 03:
Did the board find that the specification discloses steps B and C?

[00:13:55] Speaker 00:
It did not and this is a point of dispute.

[00:13:58] Speaker 00:
It says pretty clearly that elements of the claims are described and it's clear what those are.

[00:14:04] Speaker 00:
It describes MPS, not random MPS as my learned brother said.

[00:14:10] Speaker 00:
It describes in a different paragraph that 30 base pairs of random sequence can identify a chromosome, but goes on in the rest of the paragraph to describe how you use that to identify sequence-specific primers and probes, in other words, for targeted use.

[00:14:27] Speaker 03:
This is much like FW- Before you go any further, that statement in column 20 about using 30 base pairs

[00:14:36] Speaker 03:
of sequence information to associate or identify a chromosome.

[00:14:41] Speaker 03:
Does that statement work in both the random sequencing context and a targeted sequencing context?

[00:14:48] Speaker 00:
It would, but the context in which that statement appears in this specification is relating to the identification and preparation of target-specific reagents.

[00:15:00] Speaker 00:
And you can see that the next article down that's discussed is about designing primers and things like that.

[00:15:08] Speaker 00:
And the point is, this is a lot like FWP v. Biogen 749, Fed Appendix 969, admittedly an unpublished case, but a guide to the path to decision here.

[00:15:22] Speaker 00:
That was one of these cases where all the words could be found somewhere in the specification.

[00:15:29] Speaker 00:
And the late claimer came in and tried to assemble them the way that the earlier, or that the patent donor had.

[00:15:37] Speaker 00:
And what the court said, what the board said and the court affirmed, you may have those words, but you didn't describe this method as a unitary whole.

[00:15:46] Speaker 00:
And that was really the first point that the board made.

[00:15:50] Speaker 01:
What are you saying is missing in the description of the method as not being unitary?

[00:15:58] Speaker 00:
What's missing is a description that the data is generated by random MPS

[00:16:06] Speaker 00:
that that data is then normalized as is required in random MBS but not in targeted MBS.

[00:16:14] Speaker 00:
Normalization is the critical step.

[00:16:16] Speaker 01:
So if we do find in the patent references to the normalization process

[00:16:22] Speaker 01:
Is that sufficient to prompt a posita to say, based on my own knowledge and skill in the art, I now have a prompt within the pattern that prompts me to know that I have to undertake this normalization step?

[00:16:36] Speaker 00:
Well, I don't believe there is any reference to a normalization step.

[00:16:40] Speaker 01:
Let's look at the balsasubramian reference.

[00:16:48] Speaker 00:
Certainly.

[00:16:51] Speaker 01:
and paragraph 71 and 72.

[00:17:00] Speaker 01:
And there it speaks about images and other information, about reducing noise.

[00:17:05] Speaker 01:
The computer program can perform optional alignment between images, extract a single molecule data, correlate the data between the images and cycles, and specify the DNA sequence from the patterns of signals.

[00:17:20] Speaker 01:
Isn't that a reference to normalization?

[00:17:23] Speaker 00:
No.

[00:17:25] Speaker 00:
The Ballasubermanian is detecting single nucleotide polymorphisms or SNPs.

[00:17:33] Speaker 00:
All he's interested in is whether what he's found differs in one nucleotide from what can be found in the genome.

[00:17:40] Speaker 00:
And this is described in Dr. Gabriel's declaration, paragraph 65 at 3717 of the appendix, that he is not counting the number of sequences for specific chromosomes or normalizing those counts based on differences in chromosome size.

[00:18:01] Speaker 00:
That's our testimony about ballosubermania.

[00:18:05] Speaker 00:
And the other point about ballosubermania, which is in paragraph 67 of our declaration,

[00:18:10] Speaker 00:
at 3998-99, she points out that Balasubramanian does it two ways.

[00:18:20] Speaker 00:
He can do it with the naked DNA, which would be random sequencing, or he can do it with PCR-amplified DNA, which is a target-specific method.

[00:18:30] Speaker 00:
So when you ask, well, why was that cited there?

[00:18:34] Speaker 00:
First of all, the sentence it's attached to just talks about four-color sequencing.

[00:18:39] Speaker 00:
Balasubramanian does describe four-color sequencing at paragraph 64 of that application.

[00:18:45] Speaker 00:
And the reader is going to think that's all it's being cited for.

[00:18:50] Speaker 00:
But it isn't any better than the reference to the aluminum machine.

[00:18:55] Speaker 00:
And the testimony was clear, the aluminum machine could be used either for targeted sequencing or for random sequencing.

[00:19:03] Speaker 00:
And the entire thrust of this application, as the court has noted, is all about targeted sequencing.

[00:19:09] Speaker 00:
Targeted sequencing and analysis step, there's an elegance to it.

[00:19:12] Speaker 00:
The analysis step can look just to this one-to-one ratio between the number

[00:19:17] Speaker 00:
of targeted sequences for one gene and the number for the other.

[00:19:21] Speaker 00:
But when you use random sequencing, you can't do that.

[00:19:25] Speaker 00:
And there's no description in this patent that that one-to-one computation method doesn't apply to some embodiments of the invention.

[00:19:34] Speaker 01:
And frankly, a person's skill... Why wouldn't a person's skill in the art come to this patent with that knowledge already?

[00:19:41] Speaker 00:
person would come to the patent and read the patent to understand what the inventor had invented and described.

[00:19:49] Speaker 01:
Do we allow at that point the personal skill in the art to fill in blanks?

[00:19:56] Speaker 00:
Not in the description.

[00:19:58] Speaker 00:
In, if the issue is one of enablement, then certainly you need not relate what people already know in terms of enablement.

[00:20:07] Speaker 00:
When the issue is written description, ARIED makes clear that's within the four corners of the patent specification.

[00:20:13] Speaker 00:
And the point here is, there, sure.

[00:20:17] Speaker 03:
But, but, but description isn't like, you know,

[00:20:21] Speaker 03:
a demand of like an engineering specification where you have every single little itty-bitty detail all laid out in punishing detail.

[00:20:31] Speaker 00:
Correct.

[00:20:31] Speaker 00:
That's not necessary.

[00:20:32] Speaker 00:
I'm not suggesting that.

[00:20:34] Speaker 00:
There were two ways that they could have communicated to the reader here that they had in mind massively parallel sequencing of random DNA.

[00:20:45] Speaker 00:
They could have and should have... To detect aneuplity.

[00:20:48] Speaker 00:
To detect aneuplity.

[00:20:49] Speaker 00:
They could have said that.

[00:20:52] Speaker 00:
Or they could have identified, when you do that, you need to normalize.

[00:20:57] Speaker 00:
In other words, a difference in the entire approach.

[00:21:00] Speaker 03:
And they claim.

[00:21:01] Speaker 03:
OK, so just so I understand, to follow up on something Judge Rehner asked earlier, if this specification, perhaps after the citation to the Bala Sabra reference, had said, now go ahead and normalize that data that you're generating,

[00:21:20] Speaker 03:
then necessarily, even though the specification still wouldn't say altogether the magic words, random MPS,

[00:21:30] Speaker 03:
by deduction, by necessary implication, it's now talking about random MPS.

[00:21:36] Speaker 03:
Is that fair to say?

[00:21:38] Speaker 00:
I think there would still be an issue as to whether this method was really described as a unitary whole.

[00:21:44] Speaker 03:
Because normalization isn't used for targeted sequencing, right?

[00:21:48] Speaker 00:
Correct.

[00:21:48] Speaker 03:
So then what else could it be talking about if the specification had talked about normalization in relation to balisobromomania?

[00:21:57] Speaker 00:
It would be a good indicator that there was a discussion that they were talking about random sequence information.

[00:22:05] Speaker 00:
All I'm suggesting is that the way the specification is written, there's nothing that causes you to pull out Bellis-Ubermanian and read it.

[00:22:12] Speaker 00:
It says C also for four-color sequencing.

[00:22:16] Speaker 00:
And one wouldn't expect you would footnote your way into a written description of an invention.

[00:22:22] Speaker 00:
And we are blessed here.

[00:22:24] Speaker 00:
We've really got some evidence in this record of what people who actually made this invention

[00:22:31] Speaker 00:
thought you needed to say.

[00:22:33] Speaker 00:
And there are two applications in the appendix.

[00:22:36] Speaker 00:
One is Lowe's application, the involved application that starts at A4013.

[00:22:42] Speaker 00:
The other is Quake's 415 patent provisional application, which starts at page 4165.

[00:22:49] Speaker 00:
And one is Lowe's application describing random MPS.

[00:22:55] Speaker 00:
And the other is Quake's first filed application describing it.

[00:22:58] Speaker 00:
And they all do three things.

[00:23:01] Speaker 00:
Low expressly discloses random MPS for aneuploidy at paragraph 110.

[00:23:08] Speaker 00:
It identifies the marked contrast between the targeted method and the random method at paragraph 112.

[00:23:14] Speaker 00:
And it identifies the need to normalize random data at paragraphs 80 through 81.

[00:23:19] Speaker 00:
And the quick provisional application similarly expressly discloses random MPS for aneuploidy at paragraphs 33 and 70.

[00:23:28] Speaker 00:
expressly describes normalization at paragraphs 28, 54, and 70, describes the important difference between random sequencing, which uses all of the data in the sample, and digital sequencing, which looks only at targeted sequences, and how that gives you more accurate results in paragraphs 10 and 42.

[00:23:49] Speaker 00:
And most importantly here, it makes specific reference to the specification that's now in the Quake 018 involved patent.

[00:23:57] Speaker 00:
And it says it describes using digital PCR for aneuploidy.

[00:24:02] Speaker 00:
There's no recognition that it disclosed anything more.

[00:24:06] Speaker 01:
And that tells... Is random, excuse me, normalization of data required when you have random sequencing?

[00:24:16] Speaker 01:
I'm sorry, is it required?

[00:24:17] Speaker 01:
Yes, is it required?

[00:24:18] Speaker 00:
To do aneuploidy counting, yes it is.

[00:24:21] Speaker 00:
You don't need it if all you're doing is looking for single nucleotide polymorphisms as Balasuk-Subermanian was.

[00:24:28] Speaker 01:
So why is this an issue coming up now?

[00:24:30] Speaker 01:
I mean, why weren't these arguments made to begin with?

[00:24:34] Speaker 01:
It seems to me that the party should have known when you presented yourself the first time to this court, and I was on that panel, the party should have made these arguments as in the alternative.

[00:24:51] Speaker 00:
With respect, the whole point was that there was nothing in this application that specifically focused on random MPS.

[00:25:01] Speaker 00:
And they said, oh, there is this Illumina.

[00:25:04] Speaker 01:
And we reversed on that, and we said it back down.

[00:25:07] Speaker 00:
And the board said, look, Illumina can be used both ways, so it isn't specifically MPS.

[00:25:12] Speaker 00:
So this argument that is now being made here that, oh, everybody would recognize the instant

[00:25:20] Speaker 00:
I said 30 base pairs of random sequence can identify a chromosome, even though it's in a paragraph talking about making reagents for targeted detection.

[00:25:34] Speaker 00:
They would immediately know everything you're supposed to do.

[00:25:36] Speaker 00:
That's the wrong issue.

[00:25:38] Speaker 00:
The issue is reading the specification as a whole, would a person of ordinary skill and the art recognize that even though they didn't describe it, didn't come right out and describe it the way everybody else did,

[00:25:48] Speaker 00:
they nonetheless had invented the method that they're now claiming.

[00:25:53] Speaker 00:
And the answer there is

[00:25:55] Speaker 00:
without acknowledging that this very elegant one-to-one comparison method that they describe doesn't work for random, that you need normalization there.

[00:26:06] Speaker 00:
Without that, the person skilled in the art would not recognize that this pile of words here is intended to be a description of random MPS.

[00:26:16] Speaker 00:
And that basically is what the board found the first time and found in more

[00:26:22] Speaker 00:
detail in accordance with the court's helpful guidance and instruction this time around.

[00:26:31] Speaker 00:
I'd like to address the question of the prior art.

[00:26:34] Speaker 00:
There is evidence here.

[00:26:35] Speaker 00:
First, the fact that a person skilled in the art might have been able to do normalization if you had told them to do it, if it were described.

[00:26:45] Speaker 00:
is not the point, and Ariadne's clear on that, that a person of ordinary skill in the art would have been able to practice the invention had it been described as not determinative of the adequacy of description.

[00:26:56] Speaker 00:
the description has to be within the four corners of the application, and it must be of something that is not something that merely renders the invention obvious.

[00:27:05] Speaker 00:
And that's really what Quake is arguing here is, oh, from what I did write, these couple of words that I had that I've now pulled together in light of what Lowe actually claimed, a person skilled in the art would be able to practice it.

[00:27:18] Speaker 00:
Now, they rely on Capon v. Escher,

[00:27:22] Speaker 00:
for this idea that, well, you can look to the prior art for these matters.

[00:27:28] Speaker 00:
Kapon dealt with not having to repeat known DNA sequences.

[00:27:32] Speaker 00:
But more importantly here, there is testimony here from Dr. Gabriel at paragraph 19 on 5761 that random MPS was not being used for diagnostic purposes in 2006 and 2007.

[00:27:47] Speaker 00:
And such diagnostic approaches were then in early development.

[00:27:53] Speaker 00:
So this prior art was not so gloriously fully developed that all that needed to be recited was, oh, go count the chromosomes.

[00:28:03] Speaker 00:
And on that point,

[00:28:05] Speaker 00:
This is not, and the argument was made here the last time too, they want to genericize the claim.

[00:28:12] Speaker 00:
Oh, it's just counting chromosomes by whatever method.

[00:28:14] Speaker 00:
That's not the claim that's here.

[00:28:16] Speaker 00:
They had presented a generic claim like that as Quake Claim 25 at Appendix 4136, and it is not in the patent.

[00:28:25] Speaker 00:
This is a very specific claim to a very specific species of aneuploidy detection, which is a species that is not described in this patent application.

[00:28:36] Speaker 01:
Thank you very much, Your Honor.

[00:28:40] Speaker 01:
Thank you.

[00:28:41] Speaker 04:
Thank you, Your Honor.

[00:28:42] Speaker 01:
You've got two minutes.

[00:28:43] Speaker 04:
Thank you.

[00:28:44] Speaker 04:
That argument was very disturbing.

[00:28:46] Speaker 04:
The gist of it was that the patent discloses targeted sequencing, not random sequencing.

[00:28:52] Speaker 04:
Judge Chen specifically asked counsel whether

[00:28:55] Speaker 04:
The section and 20 that I pointed out before, three regarding the only 30 base pairs of random information, and said that that was a finding of targeted sequencing.

[00:29:04] Speaker 04:
You asked him if it could be used in targeted, and he told you that the board found that that was for targeted sequencing for primer making.

[00:29:11] Speaker 04:
Here, at appendix 14 at 24, passage B expressly recites random sequence information.

[00:29:17] Speaker 04:
Accordingly, even if the references that follow are not relevant to random sequence information, we find that passage B does expressly describe random sequencing.

[00:29:26] Speaker 04:
So the board found that 19 and 20 disclosed random sequencing.

[00:29:32] Speaker 04:
That was most of the argument you had.

[00:29:34] Speaker 04:
Now, on normalization, let me be clear about one thing.

[00:29:36] Speaker 04:
For any kind of down syndrome test, whether it be targeted, any way you do it,

[00:29:41] Speaker 04:
When you're taking one-to-one, you still have to clean up the data.

[00:29:44] Speaker 04:
It's not like, I mean, normalization is just one thing you do to clean up the data.

[00:29:48] Speaker 04:
It's not like you just, in targeted sequencing, you're just done.

[00:29:51] Speaker 04:
So there's no written description of that either because you don't go through all the natural cleanup you do.

[00:29:57] Speaker 04:
Because people know, now as to normalization, two critical points.

[00:30:01] Speaker 01:
What?

[00:30:01] Speaker 01:
Where's the cleaning up described in claim one of 18-Penn?

[00:30:05] Speaker 01:
I'm sorry, what's that?

[00:30:06] Speaker 01:
Where's the cleaning up described?

[00:30:09] Speaker 04:
It just says you compare one to the other.

[00:30:10] Speaker 04:
It's an enablement.

[00:30:11] Speaker 01:
That's not cleaning up.

[00:30:12] Speaker 01:
I mean, you're talking about something else.

[00:30:15] Speaker 04:
You compare data from one to the other, and you have to clean it up.

[00:30:18] Speaker 01:
But that's what's missing here.

[00:30:20] Speaker 04:
It's missing for every single thing.

[00:30:21] Speaker 04:
Then the entire patent is not written described.

[00:30:24] Speaker 04:
Because every time you do the statistical analysis, you obviously aren't just putting two numbers out there.

[00:30:29] Speaker 04:
But the point I want to make is their effort, and just indulge me in this, two quick points.

[00:30:35] Speaker 04:
Their expert admitted that one skilled in the art would know how to take into account the size of the chromosome.

[00:30:41] Speaker 04:
This is at 15981.

[00:30:43] Speaker 04:
But this was certainly well known at the time.

[00:30:45] Speaker 04:
How you do that, right?

[00:30:47] Speaker 04:
In February of 2007.

[00:30:49] Speaker 04:
To take into account the size of the chromosome and to create a normalized frequency with that information, I think someone could have done that, yes.

[00:30:57] Speaker 04:
So people would know how to do it.

[00:30:58] Speaker 04:
It's a detail.

[00:30:59] Speaker 04:
They call it a detail.

[00:31:01] Speaker 04:
So they would know it.

[00:31:02] Speaker 01:
You're out of time.

[00:31:03] Speaker 01:
Can you conclude?

[00:31:04] Speaker 04:
Yes, just the second piece to that is it's undisputed that they would know that you need to do that.

[00:31:10] Speaker 04:
The other side repeatedly said this, and I'll just give you the sites, A5768 and A4010.

[00:31:16] Speaker 04:
Their own experts said, a person skilled in the art confronting this would know that you couldn't use one-to-one because you had to normalize.

[00:31:22] Speaker 04:
Well, they would know you have to do that.

[00:31:23] Speaker 04:
They would know how to do it.

[00:31:25] Speaker 04:
That's not a wish or a plan, anything like Santa Cora or Lilly.

[00:31:29] Speaker 04:
It's a misapplication of Ariadne.

[00:31:30] Speaker 04:
Thank you very much.

[00:31:31] Speaker 01:
We thank the party for the arguments.

[00:31:34] Speaker 01:
This court is now in recess.