Erik Zhivkoplias

Marine genetics play an essential role in biodiversity studies, ecosystem monitoring, and biotechnology, yet little is known about how marine genetic sequences are translated into commercial applications. Although international frameworks such as the BBNJ Agreement and the Convention on Biological Diversity address access and benefit sharing, effective implementation requires empirical insight into how marine genes are used. Here we analyze the Marine Bioprospecting Patent Database, a global catalog of marine-derived genes appearing in patents, by combining structural topic modeling and large language model-assisted classification of 1,676 patent texts to map technological themes and actor contributions across healthcare, agriculture, industrial, and foundational biotechnology. Private-sector filings prioritize industrial and agricultural applications, whereas research institutes and governmental actors disproportionately contribute foundational, cross-cutting tools that enable downstream innovation. A curated subset of 221 patent claims mentioning marine enzymes illustrates applications ranging from bioenergy and green chemistry feedstocks to diagnostics and therapeutics. While many inventions align with sustainability objectives, others deliver incremental gains or support applications with ambiguous, if not adverse, environmental implications. Targeted policy, funding, and stewardship are needed to maximize net sustainability benefits and ensure equitable benefit-sharing under international frameworks.

Marine genetics play an essential role in biodiversity studies, ecosystem monitoring, and biotechnology, yet little is known about how marine genetic sequences are translated into commercial applications. Although international frameworks such as the BBNJ Agreement and the Convention on Biological Diversity address access and benefit sharing, effective implementation requires empirical insight into how marine genes are used. Here we analyze the Marine Bioprospecting Patent Database, a global catalog of marine-derived genes appearing in patents, by combining structural topic modeling and large language model-assisted classification of 1,676 patent texts to map technological themes and actor contributions across healthcare, agriculture, industrial, and foundational biotechnology. Private-sector filings prioritize industrial and agricultural applications, whereas research institutes and governmental actors disproportionately contribute foundational, cross-cutting tools that enable downstream innovation. A curated subset of 221 patent claims mentioning marine enzymes illustrates applications ranging from bioenergy and green chemistry feedstocks to diagnostics and therapeutics. While many inventions align with sustainability objectives, others deliver incremental gains or support applications with ambiguous, if not adverse, environmental implications. Targeted policy, funding, and stewardship are needed to maximize net sustainability benefits and ensure equitable benefit-sharing under international frameworks.

Marine genetic resources (MGRs) are crucial for understanding marine biodiversity and tracking changes in ecosystems, while also providing a valuable foundation for biotechnological innovation. The recent adoption of the UN Treaty on Biodiversity Beyond National Jurisdiction (BBNJ Treaty) and the inclusion of digital sequence information in this legally binding agreement open new opportunities for protecting biodiversity in the high seas, including the deep sea. However, a significant knowledge gap exists in fully understanding current and potential future applications of MGRs in biotechnological innovation, creating uncertainties around the Treaty's implementation. Classifying MGR functions for specific technological uses - such as antibiotic resistance, bioremediation, and biofuels - can significantly increase our understanding of the economic value of marine ecosystems and help discover new genes and enzymes with practical applications that hold great potential for sustainable development.Here, we analyzed the Marine Bioprospecting Patent Database (https://mabpat.shinyapps.io/main/), a global catalogue of marine genes used in innovation, that includes 104,467 nucleotide sequences from 1,639 marine species across 4,779 unique patents. We used technological codes, structured topic modeling, and a suite of knowledge diversity indices to evaluate the unique contributions of marine sequences to the biotechnology sector and other global economic areas. Our findings show that most MGRs are used in molecular biology to develop new enzymes and fluorescence methods, as well as in public health to create new therapeutic and diagnostic products. Additionally, we encounter various connections between marine bioprospecting inventions and the Sustainable Development Goals, emphasizing the potential of MGR utilization to promote sustainable development. Our study provides policymakers with new insight into the scope and scale of innovation based on MGR, enabling science-based decision-making related to the conservation and sustainable use of marine ecosystems.

A central barrier to effective governance and accountability in fisheries is the limited transparency of corporate ownership. Transshipment—the transfer of catches, fuel, parts, or crew between fishing and cargo vessels known as reefers—is often criticized for its opacity and poor governance. Better insight into the beneficial ownership of vessels involved in transshipment and their operational patterns could lead to more effective management. Our study presents a publicly accessible database of reefers’ owners, operators, and flags. We identified 569 individual reefers and found that Russian and Chinese owners control 26 and 20% of the global reefer fleet, respectively. Results also show that 65% of all reefer vessels fly the flags of Russia, Panama, or China. This high level of consolidation suggests considerable leverage for enhancing transparency and governance. Our findings highlight the potential for reforming existing transshipment practices through collaboration among owners, flag states, fishery regulators, and scientists.

Marine bioprospecting, which involves the exploration of genetic and biochemical material from marine organisms, can be used towards addressing a broad range of public and environmental health applications such as disease treatment, diagnostics and bioremediation. Marine genetic resources are important reservoirs for such bioprospecting efforts; however, the extent to which they are used commercially for natural product discovery and the marine sources from which they are derived are not well understood. Here we introduce a comprehensive database of marine genes referenced in patent filings, the Marine Bioprospecting Patent database. It includes 92,550 protein-coding sequences associated with 4,779 patent filings, identified by analysing all relevant records from genetic sequence databases. Three companies alone—BASF, IFF and DuPont—included sequences from 949 species (more than half of referenced species with identified marine origin). Microbial life in the deep sea, a vast and remote biome predominantly beyond national jurisdiction, is already attracting substantial economic interest; the top ten patent holders have all filed marine gene patents referencing sequences from deep-sea life. Our findings provide an updated understanding of the marine bioprospecting landscape, contribute to the sustainable use of marine biodiversity and underscore the need for policymakers to ensure stewardship of deep-sea ecosystems.

A nucleus-like structure composed of phage-encoded proteins and containing replicating viral DNA is formed in Pseudomonas aeruginosa cells infected by jumbo bacteriophage phiKZ. The PhiKZ genes are transcribed independently from host RNA polymerase (RNAP) by two RNAPs encoded by the phage. The virion RNAP (vRNAP) transcribes early viral genes and must be injected into the cell with phage DNA. The non-virion RNAP (nvRNAP) is composed of early gene products and transcribes late viral genes. In this work, the dynamics of phage RNAPs localization during phage phiKZ infection were studied. We provide direct evidence of PhiKZ vRNAP injection in infected cells and show that it is excluded from the phage nucleus. The nvRNAP is synthesized shortly after the onset of infection and localizes in the nucleus. We propose that spatial separation of two phage RNAPs allows coordinated expression of phage genes belonging to different temporal classes.

Perceptions that marine bioprospecting will deliver vast commercial benefits have placed ‘marine genetic resources’ at the center of key policy processes yet our knowledge about their importance remains limited. Here, we introduce a novel global database of marine gene sequences referenced in patent filings, the MArine Bioprospecting PATent (MABPAT) Database. It includes 25,682 sequences from 1,092 marine species associated with 3,258 patent filings, identified by analyzing all relevant sequencerecords from INSDC. Microbial life in the deep sea, a vast and remote biome predominantly beyond national jurisdiction, is already attracting significant commercial interest; all of the top 10 patent holders have filed marine gene patents referencing sequences from deep-sea life, and only three companies, BASF, IFF, and DuPont, included sequences from nearly two-thirds of all species. Our findings underscore the need for policymakers to ensure stewardship of deep-sea ecosystems while providing the most updated understanding of the marine bioprospecting landscape.

A nucleus-like structure composed of phage-encoded proteins and containing replicating viral DNA is formed in Pseudomonas aeruginosa cells infected by jumbo bacteriophage phiKZ. The PhiKZ genes are transcribed independently from host RNA polymerase (RNAP) by two RNAPs encoded by the phage. The virion RNAP (vRNAP) transcribes early viral genes and must be injected into the cell with phage DNA. The non-virion RNAP (nvRNAP) is composed of early genes products and transcribes late viral genes. In this work, the dynamics of phage RNAPs localization during phage phiKZ infection was studied. We provide direct evidence of PhiKZ vRNAP injection in infected cells and show that it is excluded from the phage nucleus. The nvRNAP is synthesized shortly after the onset of infection and localizes in the nucleus. We propose that spatial separation of two phage RNAPs allows coordinated expression of phage genes belonging to different temporal classes.