In biopatent practice, “serial identity” is a frequent pain point during the drafting and examination response stages of claims. Broad identities are prone to scrutiny that “cannot be supported by the manual,” while the scope of protection is too narrow and easily evaded by competitors through small amounts of amino acids or base mutations.

This article combines typical review practices and practical experience to systematically sort out the causes of the sequence identity dilemma and proposes full-cycle response strategies from application drafting and review responses to infringement rights protection.

---

1. The Dilemma of Sequence Identity Limitation

In the field of biotechnology, patent protection for protein or nucleic acid inventions is often written using the “sequence identity + functional limitation” approach.
For example: “A protein with Y function that shares at least 60% sequence identity with SEQ ID No.1.” ”

The intention behind this approach is clear—the object of protection is not a specific sequence, but a family of sequences with identical functions and similar structures. By setting an identity threshold (such as 60%), it is possible to include homologs in nature that differ from known sequences but retain function, as well as variants that retain core functions after artificial modification.
However, in practice, such claims often fall into a double dilemma.
During the substantive review stage, broad restrictions like 60% for identity ≥ are usually difficult to grant. The core concerns of the examiners focus on two points:

1. First, the protection range is too large, making it difficult for manual data to support the experiments. The function of proteins is determined by the three-dimensional conformation formed by folding amino acid sequences—changes in a single amino acid can affect the folding method or the conformation of the active center, causing partial or even complete loss of function. If the instructions only verify the function of wild-type or a few mutants, it is insufficient to cover a broad range of many variants that have not been experimentally validated.
2. Second, the variant features lack predictability. Based on general technical knowledge prior to the application date, those skilled in the art can easily foresee without creative effort: any sequence with a specific identity ratio (e.g., over 60%) with SEQ ID No.1 can retain equivalent functions. Homologous sequences from different species origins may have structural differences in key functional domains, and the functional effects of artificial variants cannot be accurately predicted based solely on sequence comparison.
   Ultimately, applicants are often forced to gradually compress the identity threshold from 60% to 95%, 98%, or even retain only the specific sequence that is 100% identical to SEQ ID No.1 before obtaining authorization.

But bigger problems arise after authorization.

When claims are compressed to 100% identical, the applicant essentially obtains an exclusive right of a “precise sequence” rather than the exclusive rights of a “class of functional molecules.” Competitors do not need to bypass core technologies; they only need to replace one or more amino acid residues at non-essential sites, or add non-functional tags like His-tag to the end of the sequence, thereby ensuring that the sequence of the accused infringing product is no longer 100% identical to the patent sequence, thereby successfully “exiting” the literal infringement determination.
The technical barriers to these modifications are extremely low and usually do not affect the core function of proteins. But it is precisely these seemingly minor changes that make it difficult for applicants to claim literal infringement. Even when attempting to invoke the equivalency principle, the restriction of 100% identity itself constitutes a strong legal implication—the applicant has narrowed the scope of protection to that precise sequence during examination, and courts often apply the equivalency principle more cautiously, even tending to believe that the applicant has waived a broader scope through the limitation, thereby greatly reducing the success rate of the equivalence claim.
Competitors can legally stand on the shoulders of your technology and seize your market with your inventions. Patent certificates may be obtained, but the market cannot hold on—every percentage of sameness sacrificed during the examination phase becomes an easy gap for competitors to break through during the rights protection phase.
This is precisely the industry’s pain point: “Authorization is easy, rights protection is difficult.”

---

2. The root causes

of the predicament for sequence patents Why are biological sequence patents particularly prone to this dilemma? Behind this lies the intertwining of two core provisions of patent law.

(1) The unit clause
Article 31, Paragraph 1 of the Patent Law stipulates that one application shall be limited to one invention. Two or more inventions that belong to a single general invention concept can be submitted as a single application. This “master invention concept” requires that each invention include “identical or corresponding specific technical features“—that is, the key parts that contribute to the prior art.
In the field of biological sequences, this requirement is particularly challenging. When applicants place homologous enzymes from different sources into a single application, or combine multiple variants modified at different sites for submission, examiners will ask: Is there a common functional structural unit between these sequences? If the common structure is disclosed in the prior art, it does not constitute a “specific technical feature”; If the differences between sequences are too large to induce common non-prior art structural units, the examiner will determine a lack of unity.

(2) Supporting Clauses
: Article 26, Paragraph 4 of the Patent Law stipulates that claims must be supported by the specification. This is the direct legal basis for the examiner’s challenge of broad limitation of identity.
In the field of biotechnology, the application of support clauses is particularly strict. Due to the lower predictability of biological sequence functions, examiners usually require applicants to provide sufficient embodiment data to demonstrate that different variants generally retain the target function within the claimed identity span. If it cannot be provided, its broad identity constraints are deemed unsupported by the specification.
Behind this clause lies the fundamental logic of patent law—”disclosure for protection.” The amount of technical content with definitive information disclosed by the applicant grants the right to exclusive protection to the corresponding extent; Generalizations beyond the scope of public content are not permitted.

(3) Functional Unpredictability
If the first two are legal constraints, the third is a scientific objective reality—in the vast majority of cases, it is impossible to accurately infer the biological functions of proteins from the primary sequence alone.
The function of proteins depends on their three-dimensional spatial conformation, and the formation of these conformations depends on the precise arrangement and folding of amino acid sequences. In this chain of “sequence → structure → function,” any change in the link can lead to changes in the final outcome. The same amino acid substitution may have no effect in protein A but may be completely inactivated in protein B. Homologous proteins with highly similar overall sequences may also behave significantly due to subtle differences in key functional domains.

---

3. From Application to Rights Protection: Full-Cycle Response Strategies

Understanding the causes of this dilemma clarifies the approach to responding: it cannot be passively limited only during the review phase, but should be systematically laid out in three stages: application drafting, review responses, and rights protection claims.

(1) Application Stage: The specification serves as the normative basis

for subsequent disputes. The quality of its writing directly determines the operational space for all subsequent steps. In the application documents, it is recommended to plan ahead from the following levels:
First, clearly distinguish between conservative and non-conservative areas. Through sequence alignment and domain analysis, the specification labels sequence regions essential for function (such as active centers and binding motifs) and non-essential regions.
For example, it can be recorded: “Amino acids from positions A to B form the active center of this protein, and conservation substitutions outside this region usually do not result in loss of function.” “This record can be used during the review phase to demonstrate the reasonable predictability of the variant’s function, and during the rights protection phase, it constitutes internal evidence in the specification that asserts the equivalent of infringement.”
Second, clearly list the types of mutations that are allowed. Based on well-known knowledge in the field and preliminary experimental data, explain which categories of modifications (conserved amino acid substitution, terminal tag fusion, signal peptide displacement, etc.) typically do not substantially affect protein function. This content can be used both to respond to examiners’ questions about “support” and to prove in infringement litigation that the alleged modifications to the product are non-substantive.
Third, provide functional validation data for multiple sequence variants as much as possible in the embodiments. Even if it is impossible to exhaust all variants, performing functional tests on homologous sequences or mutations from different loci of different species origins can provide key evidence supporting broad identity claims in subsequent reviews.

(2) Examination Stage: Construction

of a Layered Claims System When examiners request to limit identity, it is recommended to adopt a layered limiting strategy rather than reducing everything to 100% at once.
Specifically, independent claims can be appropriately narrowed to a range acceptable to the examiner (such as 95% or 98%) to ensure the smooth progress of the examination process. However, variants with lower identical ratios (such as 90%, 85%, 80%), variants that specify specific functional domains, and different uses and method claims must be fully retained in the dependent claims.
Here is an example:

• Independent claim 1: A protein with at least 95% identity with SEQ ID No.1;
• Dependent claim 2: A protein with at least 90% identity with SEQ ID No.1;
• Dependent claim 3: A protein with at least 85% identity with SEQ ID No.1;
• Dependent claim 4: A protein that is at least 80% identical to SEQ ID No.1.

At the same time, independent dependent claims are written separately for different functional domain variants, and method claims are written separately for different uses. Thus, even if independent claims are compressed, subordinate claims still retain multiple layers of protection.

It should be noted that after passing the examination, the broad scope reserved in the dependent claims actually exists in the authorization announcement text and can serve as a basis for asserting rights in infringement litigation—as long as the accused infringing product falls within the scope of protection of the dependent claims, it constitutes infringement. Therefore, independent claims can be compromised, but the depth of dependent claims must be retained.

(3) Rights Protection Stage: Systematic Application
of the Equivalence Principle
When a third party evades the issue through non-essential loci mutations or terminal label addition, applicants should not be in a hurry to abandon their claims simply because the sequence is not 100% identical. The principle of equivalency holds key applicability in such cases.
According to the “Several Provisions of the Supreme People’s Court on Issues Concerning the Application of Law in the Trial of Patent Dispute Cases,” equivalence refers to characteristics where, by essentially the same means, essentially the same functions and effects are achieved, and that ordinary technical personnel in the art can associate the alleged infringement without creative effort.
In litigation, it is recommended to build an argument system around three levels:
Level One: Internal evidence in the specification. Citing the division between conservative and non-conservative zones recorded in the specification during the application stage proves that the mutation site of the accused infringing product lies in the non-essential area clearly marked by the patent holder.
Level Two: External technical evidence. Through expert testimony or well-known technical literature, it was demonstrated that this site is indeed a non-essential region and that the mutation does not alter the core function of the protein.
Third level: The claim of the substance of the invention. Clarified to the court: the so-called “circumvention” by the accused infringer is not a genuine technical bypass but a minor adjustment based on the structure-function relationship information disclosed by the patent holder. This kind of behavior of using patent disclosed content to implement non-substantive modifications is precisely the object of regulation intended by the principle of equivalence system.

---

4. Conclusion

The identity dispute over  biological sequence patents always navigates the balance between licensing boundaries and protection value. What seems like a game of numerical proportional examination is essentially a precise grasp of the structural and functional logic of biotechnology, the principle of patent publication, and the boundaries of innovation protection.

For the biopharmaceutical field, high-quality sequence patent protection has never been static locking to a single sequence, but rather dynamic safeguarding of a class of innovative functional systems. Only by establishing a full-cycle, systematic patent layout mindset can technological innovation truly take root, serve as a solid intellectual property barrier, and safeguard biotechnology innovation.

---

MeritsIP: Your Global Intellectual Property Partner

At MeritsIP, we provide comprehensive IP services in biomedicine, medical devices, manufacturing, semiconductors, and AI. We support 200+ top clients globally with strategic, high-quality IP services.

Stay Connected

• • Website: https://www.meritsipglobal.com

• • LinkedIn: Follow us on LinkedIn

• • Email: meritsip@meritsandtree.com

© 2025 MeritsIP. All Rights Reserved.