Invitae Corp v. Natera Inc

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Invitae Corporation (Delaware)
  • Defendant: Natera, Inc. (Delaware)
  • Plaintiff’s Counsel: Farnan LLP; Weil, Gotshal & Manges LLP
• Case Identification: 1:21-cv-00669, D. Del., 05/07/2021
• Venue Allegations: Venue is alleged to be proper in the District of Delaware because Defendant is a Delaware corporation and has previously initiated litigation in the district, making it a convenient forum.
• Core Dispute: Plaintiff alleges that Defendant’s Signatera minimal residual disease test infringes a patent related to methods for assembling DNA sequence reads to identify genetic mutations.
• Technical Context: The lawsuit concerns computational methods used in next-generation sequencing (NGS) to improve the accuracy of identifying genetic variations, a critical component of modern genomic medicine and diagnostics.
• Key Procedural History: The patent-in-suit is a continuation of prior applications. The complaint notes that Defendant Natera has previously initiated unrelated litigation in the District of Delaware, which Plaintiff cites to support jurisdiction and venue. The complaint also preemptively addresses patent eligibility under 35 U.S.C. § 101, suggesting an anticipated defense.

Case Timeline

Date	Event
2012-04-04	’799 Patent Priority Date
2017-08	Natera launches commercial Signatera test
2019-05	Natera begins offering Signatera as a clinical laboratory developed test
2020-03-31	U.S. Patent No. 10,604,799 Issues
2021-05-07	Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

A. U.S. Patent No. 10,604,799 - "Sequence Assembly" (issued Mar. 31, 2020)

The Invention Explained

• Problem Addressed: The patent describes prior art DNA sequencing analysis as suffering from a "tradeoff" between detecting different types of mutations ('799 Patent, col. 1:62-67). Specifically, techniques struggled with positional accuracy for short reads, the detection of long insertions/deletions ("indels"), and the correct interpretation of mutations located near indels (Compl. ¶13-14; ’799 Patent, col. 2:7-26). Existing methods either sacrificed sensitivity for one type of mutation to gain it for another, or were computationally intractable for large datasets (Compl. ¶13; ’799 Patent, col. 2:3-6).
• The Patented Solution: The invention proposes a multi-stage computational method to improve accuracy. First, short DNA sequence "reads" are assembled into longer, intermediate sequences called "contigs." Second, these contigs are aligned to a known reference genome to establish their general location and identify larger-scale variations. Third, the original short reads are aligned back to their respective contigs. Finally, the information from both the "contig-to-reference" alignment and the "read-to-contig" alignment is combined to produce a final, highly accurate map of mutations relative to the reference genome ('799 Patent, Abstract; col. 4:38-57). This two-tiered alignment process is intended to leverage the strengths of both contig-based and read-based analysis ('799 Patent, col. 2:53-58).
• Technical Importance: This approach claims to provide the positional accuracy of a contig-based alignment while retaining the detailed information from each individual read, thereby overcoming the sensitivity tradeoffs inherent in prior art methods (Compl. ¶18; ’799 Patent, col. 2:53-58).

Key Claims at a Glance

• The complaint asserts infringement of at least independent claim 1 and dependent claim 5 (Compl. ¶22).
• Independent Claim 1 recites a method with the following essential elements:
  • obtaining a sample comprising template nucleic acid;
  • sequencing the template nucleic acid to generate a plurality of sequence reads;
  • inputting a reference genome and the sequence reads into a computer system to perform the steps of:
    • assembling a contig from at least some of the sequence reads;
    • identifying contig:reference descriptions of mutations by aligning the contig to the reference genome;
    • identifying read:contig descriptions by aligning each of the sequence reads to the contig; and
    • combining the contig:reference descriptions with the read:contig descriptions to produce read:reference descriptions to map positional information of mutations.
• The complaint reserves the right to assert other claims (Compl. ¶22).

III. The Accused Instrumentality

A. Product Identification

The accused instrumentality is Natera's "Signatera" test, described as a commercial liquid biopsy test for cancer diagnostics and a laboratory-developed test for detecting molecular residual disease (Compl. ¶5, ¶19-20).

B. Functionality and Market Context

The complaint alleges that the Signatera test involves collecting tissue and blood samples from patients, performing whole exome sequencing, and analyzing the resulting data to identify somatic mutations (Compl. ¶20-21). This analysis is allegedly used to generate a custom PCR panel for monitoring residual disease (Compl. ¶20). Plaintiff alleges, on information and belief, that Natera performs this test at its facility in San Carlos, California, and that the underlying analysis uses tools like the Genome Analysis Toolkit (“GATK”) HaplotypeCaller or a similar method (Compl. ¶5, ¶21).

No probative visual evidence provided in complaint.

IV. Analysis of Infringement Allegations

The complaint references a preliminary claim chart in Exhibit 6, which is not attached to the publicly filed document (Compl. ¶22). In lieu of a table, the infringement theory is summarized below based on the complaint's narrative allegations.

The complaint alleges that Natera’s performance of the Signatera test infringes at least claim 1 of the ’799 Patent, either literally or under the doctrine of equivalents (Compl. ¶21). The core of the infringement theory is that the computational analysis underlying the Signatera test meets each step of the claimed method. Plaintiff alleges that Natera's process involves sequencing a patient's DNA to generate reads, and then using analytical tools (such as GATK HaplotypeCaller or Mutect2) to identify somatic mutations by comparing tumor tissue sequence with matched whole blood sequence (Compl. ¶21). Plaintiff's theory appears to be that these software tools necessarily perform the claimed steps of assembling contigs, aligning them to a reference, aligning reads back to the contigs, and combining the results to map mutations, thereby practicing the patented method (Compl. ¶21, ¶26).

• Identified Points of Contention:
  • Scope Questions: The complaint preemptively argues that the claimed invention is not an abstract idea but a "concrete technique" tied to DNA sequencing technology (Compl. ¶16-18). This suggests a central dispute will be whether the claims are patent-eligible under 35 U.S.C. § 101, or if they claim an abstract mental process or mathematical algorithm applied to generic computer components.
  • Technical Questions: A key factual question will be whether the accused GATK HaplotypeCaller, Mutect2, or other Natera software actually performs the specific, multi-step process recited in claim 1. The dispute may focus on whether Natera's process truly involves the sequential steps of "contig-to-reference" alignment followed by a separate "read-to-contig" alignment and then a "combining" step, or if it employs a different, more integrated analytical model that does not map onto the claim elements.

V. Key Claim Terms for Construction

• The Term: "contig"

• Context and Importance: The concept of a "contig" is foundational to the claim. The patent describes it as being assembled from sequence reads and as a "consensus sequence" ('799 Patent, col. 4:47-51). The specific structural and data requirements for a set of data to be considered a "contig" will be critical. Practitioners may focus on this term because the infringement analysis depends on whether the intermediate data structures generated by Natera’s accused software meet the claim's definition of a "contig."

• Intrinsic Evidence for Interpretation:

  • Evidence for a Broader Interpretation: The specification suggests a contig can be formed by various methods, including "a greedy algorithm" or "de novo assembly," and can simply represent "the relationship between or among a plurality of segments of nucleic acid sequences" ('799 Patent, col. 3:4-6; col. 12:63-67). This could support a broad definition covering many forms of intermediate sequence assemblies.
  • Evidence for a Narrower Interpretation: The patent also describes a contig as a "consensus sequence, i.e., a probable interpretation of the sequence of the nucleic acid represented by that contig" ('799 Patent, col. 4:47-51). This language could be used to argue for a narrower construction requiring a specific type of processed, consensus sequence, not just any temporary grouping of reads.

• The Term: "combining the contig:reference descriptions with the read:contig descriptions"

• Context and Importance: This final step defines the novel output of the claimed method. The entire purpose of the invention is to use this "combination" to overcome the deficiencies of prior art methods ('799 Patent, col. 4:52-57). The case may turn on whether Natera's process performs an equivalent "combining" step or achieves a similar result through a distinct, non-infringing process.

• Intrinsic Evidence for Interpretation:

  • Evidence for a Broader Interpretation: The patent describes the outcome of this step functionally: "to map positional information of mutations found in the individual reads relative to the reference" (Compl. ¶15; ’799 Patent, col. 22:4-7). An argument could be made that any process that uses both contig-level and read-level alignment data to produce a final mutation map performs this "combining" step.
  • Evidence for a Narrower Interpretation: The specification describes this as a discrete step where the "output of the local alignment... can be combined with the output of the reference alignment" ('799 Patent, col. 4:38-42). This suggests a specific sequence of operations involving two distinct data outputs that are subsequently merged. An argument could be made that a process not following this two-output structure does not infringe.

VI. Other Allegations

• Willful Infringement: The complaint does not contain an explicit allegation of willful infringement. It does, however, request a determination that the case is "exceptional" under 35 U.S.C. § 285 to support an award of attorneys' fees (Compl., Prayer for Relief ¶D).

VII. Analyst’s Conclusion: Key Questions for the Case

This dispute appears to center on three fundamental questions for the court:

1. Patent Eligibility: Is the claimed method a patent-ineligible abstract idea or mathematical algorithm under 35 U.S.C. § 101, as Natera will likely argue, or is it a specific, concrete improvement to computer-implemented DNA sequencing technology, as Invitae contends?
2. Definitional Scope: What are the precise technical requirements for a data structure to be a "contig" and for a process to constitute "combining" the two types of alignment descriptions as claimed? The construction of these terms will define the boundaries of infringement.
3. Factual Infringement: As a technical matter, does Natera's Signatera test—and specifically its alleged use of tools like GATK HaplotypeCaller—actually perform the specific, multi-stage alignment process recited in claim 1, or does it utilize a fundamentally different bioinformatic approach to mutation detection?