Bioinformatics: Exploiting Biological Data for Patenting in India

Bioinformatics: Exploiting Biological Data for Patenting in IndiaIn an era of growing dependence on digital data, the availability of large biological data opens a new arena of innovative activity through Bioinformatics and computational biology. The term ‘bioinformatics’ found references in the writings of Paulien Hogeweg and Ben Hesper in 1970, where they refer to the study of information processes in biotic systems. It is not the study of raw biological data contained in DNA, RNA, and protein sequences, but it analyses and explores the possibilities of its applications to produce meaningful information.

For commercial purposes, the bioinformatics product market can be divided into four segments: databases, software, hardware, and custom consulting. When we talk about bioinformatic tools, we find that bioinformatics hardware is not different from traditional computer hardware as it comprises servers, desktop computers, and other storage devices, but with a major difference that it is typically designed with faster processing ability and greater storage capacity to process to store, organise, integrate, analyse, and/or distribute enormous amount of biological data. When we look into the bioinformatic products databases, we come across raw biological data contained in DNA, RNA, and protein sequences of the unicellular and higher living things, including human beings.

Bioinformatics: Content and Coverages

If we look at the official definition of ‘Bioinformatics’ as given by the National Centre for Biotechnology Information (NCBI) USA, it states that “Bioinformatics is the field of science in which biology, computer science and information technology merge into a single discipline. There are three important sub-disciplines within Bioinformatics: the development of new algorithms and statistics with which to assess relationships among members of large data sets; the analysis and interpretation of various types of data, including nucleotide and amino acid sequences, protein domains and protein structure; and the development and implementation of tools that enable efficient access and management of different types of information”.

Research and inventing activities in bioinformatic fields revolve more or less around these three fields, and patenting activity is not limited to generating biological information databases. It extends to designed and tailor-made algorithms to meet the requirement of faster processing capability and greater storage capacity. Does it mean bioinformatics inventions are mainly for software tools and screening methods?

Patentable Bioinformatics Inventions!

A closer look at the patent database reveals that system and method patents for (a) predicting gene expression using deep learning, (b) identifying genetic variants associated with diseases using machine learning, (c) designing synthetic biological pathways using artificial intelligence,(d) analysing genomic data using neural networks and (e) predicting protein structure using deep learning are possible patentability options availed with inventors working in this emerging field thanks to successful collaboration between biotech and infotech companies. This means that it is possible to obtain patents for claims such as (a) a method of finding gene relationship, (b) an apparatus drafted as computer programmed to……., (c) means-plus-function, (d) products such as kits, assay, API, (e) User interphase to accept command, (f) data structure for storage of retrievable biological data, (g) computer-readable medium drafted as Beauregard claim.

Patenting Biological Databases?

Biological Databases are the “heart” of bioinformatics. The amalgamation of biotech and infotech processes and methods leads to the accumulation of biological data that is accessible to bioinformatic software for sorting, integrating, analysing, and distributing database contents. This bioinformatic segment is dominated by data mining software that has the capability to search raw biological information databases to find relationships, patterns, and functions between biological components. Bioinformatics hardware is not different from the traditional computer hardware such as servers, desktop computers, and other storage devices.

However, it is designed to meet the requirement of faster processing capability and is tailor-made to enhance its storage capacity. This activity requires writing and running software programs that use algorithms from graph theory, artificial intelligence, soft computing, data mining, image processing, and computer simulation. It is not surprising to find that bioinformatics algorithms per se are not considered patentable subject matter in almost all the patenting jurisdictions, and so are the computer programmes per se. However, copyright protection for biological databases is available in India under Section 2(o) of the Copyright Act as amended in 2002, which defines ‘literary work’ and includes computer programs, tables, and compilations, including computer databases (including biological databases). This protection matches copyright protection granted in the US and Europe. Nevertheless, this leaves us with a pertinent question: What is the possible category of bioinformatic invention that can be protected under patent law?

What is not Considered Patentable?

The first patentability hitch faced by bioinformatic-related inventions is that they are considered akin to mental processes. It is understood that the activity that can be done mentally is non-patentable subject matter. Another hurdle is that such inventions must overcome the examiner’s bias towards typical regular-type software claims. One aspect of bioinformatic-related patent applications is attributed to overzealous inventors who seek broad claims without identifying the practical application of the invention. In India, Section 2(1)(J) typically mandates that the invention claimed is new, inventive, and has an industrial application. Additionally, section 3 (k) says that “mathematical or business method or a computer programme per se or algorithms are not patentable invention in India. Under section 3(m), “a mere scheme or rule or method of performing mental act or method of playing games” is not patentable.

Position in the UK

Indexing was held as non-patentable in Wards Appln. (1912) 29 RPC 79 where claim for ‘System of indexing’ was refused by the examiner. In Wards matter, James Davis Ward applied for a Patent for “Improvements in or connected with book and like indexes”, with claims comprised of a certain method or system of indexing and means, consisting of sheets with certain matter printed thereon, for carrying out the method or system. The patent examiner refused the application as it was only a method of indexing. The Solicitor General (SG) who heard this appeal found this system was similar to the fingerprint system developed by Sir Edward Henry while he was in India. In this procedure, he would record the suspected person’s fingerprint on paper when arrested. The real difficulty in all these systems is actually to find the thing you want. That is, you want to find a system of indexing. The SG observed, “There is an ingenious system used in connection with the fingerprint system by which you can discover whether the man before you is one of the thousands of persons whose fingerprints you have already had. That is a system by which you can find the place in a list, and I think the suggestion here is of a similar kind.”

The SG ruled that in Ward’s application, indexing was based on the fact that “there are only so many letters in the alphabet, and by his system, he claims to be able to find the place of a particular person more rapidly than otherwise, but that does not depend on the fact that he has manufactured or invented anything; it depends on the fact that he has hit on an idea which, applied to this branch of commerce, will save time because of the circumstance that every name is made up of a succession of letters”.

The SG upheld the refusal and ruled that “I am of opinion that the ingenious person who has suggested this arrangement, though he may call it an invention for some purposes, cannot get a Patent for it.”

Similarly, in R’S Appln. (1923) 40 RPC 465, the applicant’s claim for the placing on the reverse side of the disc, which contains in the middle a printed label and on the circumference a record of spoken words, was refused by the examiner and on appeal the SG found “this is a mere system or a mere method of arranging words -or objects upon the discs, and that I think, is not, a manner of new manufacture …… hence it is not considered as a patentable subject matter”.

Position in the US

Looking at the more recent cases, we find that even in the US, bioinformatics patent applications face a Section 101 rejection. For example, In re Board of Trustees of Leland Stanford Junior University, [No. 20-1288 (Fed. Cir. 2021)], the Federal Circuit affirmed Patent Trial and Appeal Board (PTAB) decisions of finding two Stanford patents directed to haplotype phasing ineligible as abstract ideas. Again, in CardioNet, LLC v. InfoBionic, Inc. [No. 20-2123 (Fed. Cir. 2021)], the CardioNet heart monitor patent was found invalid by the Federal Circuit for not citing an inventive concept. If the former situation was difficult to reconcile, the later patent refusal could have been saved by cautious drafting skills.

Exploring Possibilities of Patenting Bioinformatics Inventions

Going forward, let’s explore all the possible avenues that can be used to find a suitable draft of the claims in the specification that is found acceptable by the patent office for the grant of a patent. Here, we may talk about the drafting techniques that may be found useful to help pre-emptively address subject matter eligibility concerns, particularly by providing as much detail as possible about the algorithm. Here, the stress should be on specifying all visible improvements made to computer functionality achieved by the claimed invention. This means linking the means with functionality through technical advancements or effects to achieve the results claimed in bioinformatic-related inventions. We may look at the representative list of patents already granted in India and other jurisdictions on bioinformatic-related patents.

Examples of Granted Patents in India and other Jurisdictions

Patent/date of grantTitleInventor /Assignee
US468320AProcess for amplifying nucleic acid sequences.Kary B. Mullis/ Roche Molecular Systems Inc
US10956714B2

23.01.2023

Method and apparatus for detecting living body, electronic device, and storage mediumBeijing Sensetime Technology Development Co Ltd
US20190367965A1

25.04.2023

Portable genetic detection and analysis system and methodIllumina Inc
US10,956,714

 

System and Method for Predicting Gene Expression Using Deep LearningStanford University

 

US10,741,329

 

Method and System for Identifying Genetic Variants Associated with Disease Using Machine LearningIllumina, Inc.

 

US10,621,111

(expiry 2038)

 

System and Method for Analysing Genomic Data Using Neural NetworksGoogle LLC

 

US9,965,688

(expiry 2034)

System and Method for Predicting Protein Structure Using Deep LearningUniversity of California

 

US10,355,233

 

Method and System for Designing Synthetic Biological Pathways Using Artificial IntelligenceMIT
IN 265493

10.03.2015

Identification of cancer protein biomarkers using proteomic techniquesYale University
IN 545836

24.07.2024

Method and system for digital biomarkers platformTata Consultancy Services Limited

 

IN 461391

23.10.2023

A microfluidic device for the POC detection of oleophilic biomarkers in hydrophilic analytesIndian Institute of Technology Guwahati
IN 384852

23.10.2023

Method for analysing pluripotent stem cell biomarkers and implementations thereof23IKIGAI PTE LTD

 

The prosecution history of the Indian Patent 545836 of Tata Consultancy Services Limited reveals that out of 32 claims, only 8 claims were allowed. The amended claims for the invitro analytical method and kit related thereto were allowed when applicants deleted claims 2, 6, and 8 (selection of algorithm), 12 (metadata) and to meet the objections under Section 3(i) (method of treatment) and 3(k) (algorithm and computer programme). Similarly, the prosecution history of IN 461391 was granted to the Indian Institute of Technology Guwahati for only 6 claims after the applicant deleted claims 8-11, which were found to fall under Section 3(i) method of treatment and objection raised by the patent office. In 23IKIGAI PTE LTD patent IN 384852, only method for analysing pluripotent stem cell biomarkers claims 1-12 were allowed, and the claims 14-18 for method of treatment and use of stem cells (27-31) were deleted by the applicant to meet objection under Sections 3(i) and 3(j) respectively.

Position in India

In addition to Section 2(1) (Ja) and Sections 3(k) and (m) limitations in the Indian context, Section 3(i) prohibits the grant of a patent for ‘any process for the medicinal ……diagnostic …. or other treatment of human beings’ and Section 3(j) put a statutory bar on patenting ‘plants animals and part thereof’. This means that mere determination of the DNA sequence is not considered patentable. However, an inventor who identifies genes and their useful function, isolates and clones them, and obtains copies of the synthetic gene is considered patentable, but with a restriction that it is not usable for diagnosis or therapy. Patents for biomarkers are allowed if they are used for in vitro analysis and do not lead to diagnosis.

Looking Forward

If we look back to see the patent eligibility of bioinformatic-related inventions, we will find that all aspects of bioinformatics are considered patentable subject matter in many jurisdictions, including India. We will also find that bioinformatics inventions can be patented with claims similar to those of other computer methods. However, patents for algorithms, mathematical methods, and computer programmes are considered non-patentable subject matter in India and elsewhere.

If, in India, Section 3(k) puts a patent legibility bar on such invention, in the USA and Europe, it is guided by the court’s precedents that limit patenting some aspects of bioinformatics-related inventions. However, the standards of bioinformatic and software patents followed in the US, Europe, and India are typically based on judicial precedents regarding computer-related and biotechnology-related inventions. Encouraging aspects for inventors in this commercially profitable bioinformatics tools as an emerging field is that the number of bioinformatics-related patents granted in the United States, the European Union, and even India is growing exponentially. The journey of bioinformatics patent applications from the first patent application in 1998 to more than 18,000 applications in 2023 is quite impressive. If we see the estimated global bioinformatics market in terms of revenue, according to one estimate, it is likely to reach $18.7 billion by 2027.

With the increasing use of artificial intelligence and machine learning in bioinformatics, the patenting activity in fields like bioinformatics methods, new algorithms for predicting disease risk, drug efficacy, cancer research, infectious disease monitoring, drug discovery, personalised medicine, etc., is likely to grow significantly. Expert opinion may be handy when the applicant seeks to obtain bioinformatic-related patents in India in view of India’s specific patent law provision on the non-patentability of certain inventions, as discussed above. It is heartening to note that many patents relating to bioinformatics-related inventions are also allowed in India to Indian and foreign applicants.

Author: DPS Parmar

First Published by: Mondaq here