From Open Source to Collaboration: AlphaFold’s Journey for an Inclusive Open Science Community
- Renaldo Pool, BHSc
- Jul 16
- 16 min read
Updated: Sep 18
AlphaFold is transforming protein structure prediction by making cutting-edge science accessible to researchers worldwide. It's an open-source platform that fuels collaboration, drives innovation across disciplines, and supports equitable progress in medicine, agriculture, and environmental science. This final article explores its evolution, global impact, and commitment to inclusive open science.
David Martin
Cindy Hamilton
Table of Contents
Introduction
With the introduction of AlphaFold to the world of structural biology, its impact has opened the door to numerous innovative solutions. Previously persistent challenges have been identified and reviewed, and this emerging AI platform offers a necessary roadmap to navigate them with accuracy and reliability. Regardless of AlphaFold’s current limitations, it continues to advance through ongoing refinement and database expansion, and its real-world applications in addressing unresolved problems in science are becoming increasingly manageable.
Furthermore, AlphaFold's open-source availability cultivates a culture of global scientific participation, communication, cooperation, and idea sharing. With over two million researchers worldwide able to access this platform, including scientists from low- and middle-income regions, another challenge is addressed: limited resources for scientific advancement in certain countries. Enabling accessibility to AlphaFold encourages scientific engagement and rapidly accelerates discovery and contributions to various scientific disciplines. Integrating multi-omics approaches expands insights for drug target discovery, therapeutic drug development, disease mechanisms, agriculture, and environmental science.
The final article in the five-part series invites readers to follow the progression of AlphaFold’s protein structure prediction from its inception to its latest available version and capabilities, the policy changes enabling global accessibility, and the effort to address equitable practices in scientific involvement. The article highlights AlphaFold’s contributions to science through its commitment to transparent, open science and innovative diversification, interdisciplinary integration across various sectors, and training initiatives designed to support scientific and technological advancement.
Core Concepts Discussed in This Article
AlphaFold’s strength lies in its open accessibility, with more than two million researchers globally able to utilize the three-dimensional structure prediction data. With shifts in 2023 relating to policy changes, AlphaFold 3—its latest iteration—enables this advanced scientific tool to accelerate scientific discoveries further.
Cross-industry collaboration is another aspect of AlphaFold's application that drives scientific advances. It brings scientists and researchers from academia, industry, and the government together to address common concerns worldwide, fostering cooperation and collaboration.
With AlphaFold’s software continuously evolving, training programs are essential. The AlphaFold Education Summit, EMBL-EBI workshops, and university training initiatives encourage participation to improve the skills associated with operating AlphaFold.
A crucial focus is ensuring accessibility to low—and middle-income countries, regions where previous scientific involvement was curbed due to resource constraints. Open access to this AI-driven structural biology tool creates a culture of inclusivity, collaboration, and equity in the research community worldwide.
AlphaFold: From Conception to Accessibility in the Scientific Community
Overview of AlphaFold’s Range of Capabilities
Since AlphaFold’s inception and availability to the scientific community, this AI tool, capable of constructing three-dimensional protein models, has seen an expansion of 300 000 protein structures on average in 2021 as part of its Protein Structure Database (AlphaFold DB), to more than 214 million predicted protein models as of 2023 (Akdel et al., 2022; Varadi et al., 2023). Indicative of the advancements made by AlphaFold in a short time span, by leveraging AI technology for scalability and real-world applications of entire proteomes (Balasubramanian, 2024).
Not only has the Protein Structure Database expanded, but AlphaFold itself has undergone significant changes, with each version succeeding far beyond the previous iterations by advancing its capabilities. Since the initial version was revealed in 2018 at the 13th CASP competition, AlphaFold2 was introduced at CASP14 in 2020, and AlphaFold3 was released in 2024. The latest version mimics interactions between proteins and other biomolecules, such as DNA, RNA, and other small molecules, such as ligands (Akdel et al., 2022; Desai et al., 2024).
Furthermore, AlphaFold has expanded beyond its initial applications, whereby it can be combined with alternative biological databases. For example, integration with ChannelsDB 2.0 adds structural information from PDB and AlphaFoldDB to highlight protein channels, tunnels, and pores. AlphaFold’s database has become a significant foundation within the broader bioinformatics community, which also includes integration with other relevant biological data sources for an extensive analysis of data, such as the Protein Data Bank (PDB), UniProt, Ensembl, InterPro, and MobiDB (Cusack et al., 2021; Varadi & Velankar, 2022; Varadi et al., 2023).
AlphaFoldDB provides highly accurate and reliable predictions of proteins involved in various aspects of life. It emphasizes the impact of revealing protein structure secrets that have been difficult to detect due to limitations experienced with conventional methods (Durairaj et al., 2023). In addition, combining AlphaFold with multi-omics further expands its application, for instance, studying various disease mechanisms for therapeutic drug target discovery and design, which is discussed later in this article.
Moreover, retrospectively examining AlphaFold's conception and how its scope of application for biological structures has branched out offers significant insight into its improvement.
Accessibility in Scientific Engagement
A significant aspect of AlphaFold’s design is ensuring accessibility to a wide range of users, including scientific researchers, industry innovators, and other participants involved in the scientific community (Varadi et al., 2023). The global impact of AlphaFold’s utilization involves over two million researchers globally, with more than 600,000 individuals residing in low and middle-income countries, highlighting the influence this deep learning resource has on democratizing high-end structural biology information and technology in sectors and countries with economic restrictions.
Even though the most recent version, AlphaFold 3, faced backlash due to accessibility limitations, a change in policy by Google DeepMind and Isomorphic Labs enabled the academic community to access the source code (Kasanmascheff, 2024). The initial release of AlphaFold 3 restricted access to protein structure information by limiting it to a closed server; however, the scientific community criticized it for stalling the drug discovery and development process, which also impacts transparency and reproducibility (Desai et al., 2024; McKay, 2024; Krishnaswamy, 2024). After the policy change in 2024, academic researchers can now access information on biological structures. This drives rapid discovery and innovation in biology and medicine, leveling accessibility to all educational institutions and researchers (McKay, 2024).
The policy's compilation highlighted that scientists affiliated with academic institutions can access open-source information only on request, which can only be used for non-commercial research (Kasanmascheff, 2024). In this way, DeepMind ensures that this AI technology is not misused through commercialization efforts (McKay, 2024; Krishnaswamy, 2024).
Data Access Mechanisms
The scientific community has an array of options for accessing the AlphaFold database, tailored for each researcher’s needs and technical capabilities. These include direct files through FTP, Google’s Cloud Public Datasets, and programmatic access endpoints. It provides different levels of user accessibility, considering each user’s computational expertise for optimal utilization of AlphaFold DB (Varadi & Velankar, 2022; Varadi et al., 2023).
Creating a Culture of Scientific Collaboration and Community
AlphaFold’s impact across various sectors is already well-established. These sectors include academia, industry, research, and government. Due to its wide range of applications, it is considered a valuable tool for promoting collaboration among key stakeholders. This section will examine how remarkably versatile it is to expand on potential future partnerships.
Firstly, AlphaFold is capable of revolutionizing research in academia. Its cutting-edge protein prediction algorithms and open-access model can support advances in medicine, public health, molecular biology, biotechnology, and bioinformatics (Desai et al., 2024). This makes it an invaluable asset for researchers and educators alike. By integrating AlphaFold into classroom lectures and academic discussions, institutions can help prepare future healthcare professionals and scientists to work more efficiently and cost-effectively. Promoting its use through educational resources is one way to align with the needs of a technologically advancing society (United Nations Development Programme, 2025).
Another aspect of interest is the doors that AlphaFold opens to discoveries and innovative approaches across various industries. By advancing research, new applications come, many of which require interdisciplinary collaboration. For instance, a pharmaceutical company conducting a clinical trial on a drug targeting specific neurotransmitters might need to partner with an IT firm to utilize AlphaFold’s AI-based capabilities effectively.
Similarly, the agricultural industry could apply AlphaFold to develop healthier, pest-resistant crops. For instance, if a fungal enzyme is found to damage a wide range of crops, AlphaFold can help researchers understand its structure to engineer more resilient plants (Kasanmascheff, 2024). This would naturally require collaboration between the agricultural and bioengineering sectors. A strong real-world example is Isomorphic Labs – an offshoot of Google DeepMind – which collaborates with pharmaceutical companies like Novartis and Eli Lilly using AlphaFold 3. These examples highlight the model’s interdisciplinary and collaborative nature (Varadi & Velankar, 2022; Durairaj et al., 2023; Desai et al., 2024).
Government agencies also stand to benefit significantly from AlphaFold’s applications in disease control, agriculture, and environmental protection (United Nations Development Programme, 2025). Agencies such as the Centers for Disease Control and Prevention (CDC), Environmental Protection Agency (EPA), National Institutes of Health (NIH), and Food and Drug Administration (FDA) all depend on innovative research. Collaborating with academic and research institutions that use AlphaFold can support investigations into public health issues, pharmaceutical development, pest resistance, and agricultural improvements (AlphaFold: Transforming Protein Research for Australian Scientists, 2025).
AlphaFold’s Integrative Applications in the Scientific Community
It is no secret that AlphaFold has already been applied to countless scientific fields, demonstrating its implication in rapidly advancing innovative approaches and discoveries across multiple disciplines (Varadi & Velankar, 2022).
Therapeutic drug discovery and personalized medicine approaches.
Research-based disease understanding and prevention.
Medical applications and interventions focused on, for instance, neuroscience and oncology.
Addressing environmental circumstances, i.e., pollution, greenhouse gas emissions, and plastic buildup prevention.
Application to agriculture.
Elucidate evolutionary biology through protein prediction.
This is an excellent framework for an inclusive and transparent research environment, highlighting AlphaFold's open science platform to promote multidisciplinary collaboration (AlphaFold: Transforming Protein Research for Australian Scientists, 2025). Researchers can replicate and validate findings in different contexts by sharing data and study results across the global scientific community (Cusack et al., 2021; Desai et al., 2024). These platforms encourage cohesion, accelerate discovery, and promote combined participation among institutions, industries, and government agencies alike.
Training Initiatives for Future Science
Given AlphaFold’s collaborative and interdisciplinary nature, training new professionals to use it effectively is essential (Varadi et al., 2023). As technology and AI rapidly advance, integrating tools like AlphaFold into the workforce can enhance productivity and innovation. Its accuracy, efficiency, and cost-effectiveness make it a powerful asset for many scientific fields.
Training researchers is especially critical. Research professionals must stay current with the latest advancements in technology and discovery. Introducing them to AlphaFold and helping them gradually integrate it into their projects can accelerate research that might otherwise be time-consuming or difficult using traditional methods (Varadi et al., 2023). Combining high accuracy with improved efficiency helps lower both cost and labor.
A concrete example of a training initiative is the AlphaFold Education Summit hosted by the European Bioinformatics Institute in January 2025. This program empowers academics and educators to explore integrating AlphaFold into their work, particularly supporting under-resourced regions with limited access. This effort represents a training strategy and a collaborative step toward global research equity (Cusack et al., 2021).
Equally important is educating researchers and professionals on the benefits of using open-source platforms. Academic institutions and workplaces can support this through training programs and workshops on open science tools (Cusack et al., 2021). Grant organizations may even consider requiring open-source engagement as part of research funding criteria. Additionally, publishers could adopt policies that require open-source sharing for research published in their journals. Together, these efforts help promote a more collaborative and forward-thinking research environment.
Additional Training Programs and Workshops for Academia and Industry Include:
OmicsBD has a 20-hour advanced curriculum on analyzing comparisons between AlphaFold 2 and 3 abilities, modeling techniques for multi-molecular complexities, and understanding prediction validations with assistance from confidence metrics.
Texas A&M’s in-person technical workshop focused on integrating different bioinformatics tools with AlphaFold DB.
Alternative resources include EBI/EMBL’s online course covering the structure prediction foundation, navigating the database, and validating predictions through experimental procedures (Cusack et al., 2021).
AlphaFold 3’s accessibility through academic affiliations provides the complete inference code made available on GitHub, model weights can be accessed through academic portal requests; however, commercial utilization is prevented for IP protection (Kasanmascheff, 2024).
The Impact of AlphaFold’s Open-Source Data on the Scientific Community
AlphaFold DB’s accessibility to the greater scientific community has profoundly influenced the world of biological structure prediction and information (Durairaj et al., 2023). Remarkable achievements have been made, whereby this AI technology’s creators were granted a Nobel Prize for this significant discovery, aiding scientific research and development. Furthermore, AlphaFold is implicated in rapidly progressing research initiatives, saving costs and time, which would take years of study for conventional methods to determine (Guo et al., 2024b). Saving time with structure prediction through this deep learning technology has enabled researchers to focus on essential aspects of the research process by freeing up resources that would otherwise have resulted in research bottlenecks.
Examples of developments include:
A group of 47 universities uses AlphaFold 3 to map previously understudied proteins from organisms referred to as extremophiles (capable of growing in extreme environmental conditions). More than 12,000 new structures from deep-ocean vents and the polar environment have been confirmed through collaboration.
Using Cryo-EM to validate AlphaFold protein predictions, different academic institutions are advancing the quality and reliability of membrane protein research, while simultaneously increasing the speed of these experiments compared to conventional methods (Cusack et al., 2021).
EMBL-EBI’s open-access training programs have ensured that over 15,000 researchers from 190 countries have received training since January 2025, providing a hands-on approach integrated with molecular dynamics simulations.
Ensuring access to AlphaFold’s software for non-commercial use has garnered worldwide attention, establishing an environment of participation where past limitations are no longer considered restricting factors (Kasanmascheff, 2024). Initiatives held by BioStruct-Africa indicate how underrepresented regions are motivated to be involved in training programs, where researchers from 15 African countries utilized AlphaFold for malaria protein studies and focused on antimicrobial resistance targets (Nji et al., 2025).
Exploring the Impact of AlphaFold in Resource-Limited Settings
The advanced protein structure prediction tool has impacted research efforts worldwide by providing global access to AlphaFold DB's source code, regardless of location or industry-related wealth. As previously highlighted, AlphaFold’s open-source initiative also allows researchers and institutions from resource-limited settings to be involved in using this AI tool. This, in turn, facilitates an environment where expensive laboratory equipment or advanced technological systems are a thing of the past, and provides equal opportunity for researchers from various economic or geographically remote settings to compete and collaborate with renowned research centers. It cultivates international inclusivity and fosters collaboration for researchers on the same level (Varadi & Velankar, 2022).
Helping Researchers Across the Globe – the Impact on Low- and Middle-Income Countries (LMICs)
A cornerstone of AlphaFold's democratizing influence is its demonstrably remarkable capacity to help researchers across the globe, particularly those in regions and institutions historically underserved by traditional structural biology infrastructure (Nji et al., 2025). The open access to cutting-edge structural prediction tools like AlphaFold 3 and vast databases of molecular models is actively bridging the global scientific divide, fostering a more equitable landscape for biological discovery, with tangible results beginning to emerge (Cusack et al., 2021).
For scientists in low- and middle-income countries (LMICs), the previously steep barriers to entry for experimental structural biology have been significantly lowered. The AlphaFold Protein Structure Database (AFDB) continues to be an invaluable starting point, and the non-commercial open-source availability of AlphaFold 3's inference code has been pivotal (Tunyasuvunakool et al., 2021; Abramson et al., 2024). We now see the fruits of this accessibility as research groups utilize these tools for impactful local research.
The implications are no longer just potential; they are increasingly realized:
Targeted Investigations of Locally Relevant Diseases: Emerging application studies from LMICs are beginning to showcase the use of AlphaFold 3 in modeling key protein complexes in drug-resistant pathogens relevant to their regions, such as malaria, helping to identify novel interaction interfaces (Nji et al., 2025). Similarly, research is underway to understand structural variations in viral proteins from regional viral isolates, with implications for refining diagnostic tools (Desai et al., 2024).
Unlocking Local Biodiversity with Structural Insights: The ability to predict structures for novel proteins is accelerating the exploration of unique biological resources (Varadi et al., 2021). Initial reports highlight instances where AlphaFold 3 has been instrumental in predicting the structures of uncharacterized enzymes from endemic medicinal plants, leading to functional annotations and the discovery of novel enzymes with potential therapeutic applications. This demonstrates a direct path from genomic sequence to structural insight and potential bioprospecting (Desai et al., 2024).
Enhanced Participation and Leadership in Global Research: Researchers from LMICs are increasingly utilizing AlphaFold-derived structural data to lead and contribute substantively to international collaborations. Reports from global health initiatives are beginning to feature the structural insights provided by partner institutions in Africa and Asia, often based on AlphaFold 3 predictions of pathogen proteins or host-pathogen interaction complexes (Nji et al., 2025).
Growth in Local Capacity and Computational Infrastructure: The availability and utility of AlphaFold 3 have spurred demonstrable growth in computational biology training and infrastructure in several LMICs. Regional bioinformatics networks actively promote structural workshops focused on AlphaFold 3 and its applications, often supported by cloud-computing grants and international partnerships, thereby building a sustainable local talent pool (Nji et al., 2025).
Accelerating Hypothesis-Driven Science: With reliable structural models more readily available, research groups are shifting resources towards experimental validation and deeper functional studies. The "AlphaFold-first" approach for target selection or variant interpretation is becoming more common, as evidenced by methodology sections in many publications from diverse geographical locations (Cusack et al., 2021).
Community-driven platforms like ColabFold (Mirdita et al., 2022) have continued to evolve, offering increasingly user-friendly interfaces and workflows incorporating aspects of AlphaFold 3's capabilities. This further democratizes access, especially for those with limited local computing power. Furthermore, developments continue to create successful cloud-based AlphaFold 3 implementation projects tailored for resource-constrained settings, further enhancing accessibility (Nji et al., 2025).
While challenges in consistent computational access, advanced training, and internet infrastructure persist in some regions, the progress since AlphaFold 3's release is undeniable (Nji et al., 2025). AlphaFold is not just distributing data; it is distributing capability, and the global scientific community is becoming more inclusive, diverse, and effective as a direct result. The "leveling of the playing field" is an ongoing process. Still, the trajectory by early 2025 shows significant positive momentum, driven by the ingenuity of researchers worldwide now equipped with these powerful predictive tools.
Ultimately, the focus on the collaborative aspect of AlphaFold’s open-source design motivates researchers to communicate, share ideas, and work together for scientific advancement (United Nations Development Programme, 2025). Through a combined effort, the scientific community can build on AlphaFold’s foundational source code, alter it for innovative applications, and combine it with new approaches, elevating the advancement of various scientific fields on different levels. It provides a sense of community and teamwork and acknowledges the collaborative aspect, where we can advance the scientific field together. This also signifies that increased participation, alongside innovative ideas, will significantly benefit society in the long term (Varadi & Velankar, 2022; Guo et al., 2024b).
Open-Source Responsibility
Lastly, the accessibility to AlphaFold’s source code also creates a culture prioritizing responsible utilization, enhanced security, and awareness of intellectual property. DeepMind’s policy stipulates and restricts specific users, highlighting the responsibility for open-source science, cultivating transparency, and ethical consideration (Krishnaswamy, 2024; Desai et al., 2024).
Conclusion
AlphaFold has expanded the limitations of conventional protein structure predictions. Despite its limitations, the progress made in scientific biology through its application is already significant. This AI tool's ability to be utilized in various fields, such as agriculture, environmental science, medicine, biology, and pharmacotherapy, stems from its open-access development.
AlphaFold will also advance as technology improves, enhancing its capabilities from understanding intricate biomolecular actions to accurately identifying disordered protein regions. It will ultimately provide significant insights into dynamic protein interactions, further accelerating scientific discovery and applications, leading to innovative solutions.
The availability of open-source software to researchers and scientists has collectively changed scientific approaches and expedited advancements. This platform affects the technicalities of AI and ML integration into science and encourages cooperation among different sectors and collaboration worldwide. In turn, it promotes global involvement, participation, and scientific discovery, addressing common issues that affect us all. From finding solutions to disease mechanisms, advancing drug development and integrative omics approaches, to creating environmentally sustainable products that will also impact agricultural sectors, AlphaFold will influence the understanding of who we are and how we came to be, providing a holistic view of life's fundamental building blocks.
Lead Researcher and Writer: Renaldo Pool, BHSc Research and Writing: David Matin, B.A. and Cindy Hamilton, MPH Review & Editing: Larrie Hamilton, MHC Visualization and Final Edit: Michelle Ryan, MHA Conceptualization and Methodology: David Priede, PhD Project Administration and Funding Acquisition: BioLife Health Research Center |
References
1. Abramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Gural, A., Hassabis, J., Jumper, J., Kelley, R., Kohl, S., Lącki, M., O'Neill, K., Siarohin, A., Tunyasuvunakool, K., & Vani, C. (2024). Accurate prediction of the structure and interaction of any biomolecular complex with AlphaFold 3. Nature. Advance online publication. https://doi.org/10.1038/s41586-024-07487-w
2. Akdel, M., Pires, D. E. V., Pardo, E. P., Jänes, J., Zalevsky, A. O., Mészáros, B., Bryant, P., Good, L. L., Laskowski, R. A., Pozzati, G., Shenoy, A., Zhu, W., Kundrotas, P., Serra, V. R., Rodrigues, C. H. M., Dunham, A. S., Burke, D., Borkakoti, N., Velankar, S., . . . Beltrao, P. (2022). A structural biology community assessment of AlphaFold2 applications. Nature Structural & Molecular Biology, 29(11), 1056–1067. https://doi.org/10.1038/s41594-022-00849-w
3. AlphaFold: transforming protein research for Australian scientists. (2025). Bioplatforms Australia. https://bioplatforms.com/alphafold-transforming-protein-research-for-australian-scientists/
4. Balasubramanian, S., MD JD. (2024, September 10). Google DeepMind and Isomorphic Labs are making rapid progress in biology and drug discovery. Forbes. https://www.forbes.com/sites/saibala/2024/09/10/googles-ai-research-arm-deepmind-is-making-rapid-progress-in-biology-and-drug-discovery/
5. Callaway, E. (2025). AlphaFold is running out of data - so drug firms are building their own version. Nature, 640(7972), 297-298. https://doi.org/10.1038/d41586-025-00868-9
6. Center for Open Science. (n.d.). What is open science? https://www.cos.io/open-science
7. Cusack, S., Eustermann, S., Kleywegt, G., Kosinski, J., Mahamid, J., Marquez, J. A., & Muller, C. (2021). Great expectations – the potential impacts of AlphaFold DB. EMBL. https://www.embl.org/news/science/alphafold-potential-impacts/
8. Desai, D., Kantliwala, S. V., Vybhavi, J., Ravi, R., Patel, H., & Patel, J. (2024). Review of AlphaFold 3: Transformative Advances in Drug Design and Therapeutics. Cureus. https://doi.org/10.7759/cureus.63646
9. Durairaj, J., Waterhouse, A. M., Mets, T., Brodiazhenko, T., Abdullah, M., Studer, G., Tauriello, G., Akdel, M., Andreeva, A., Bateman, A., Tenson, T., Hauryliuk, V., Schwede, T., & Pereira, J. (2023). Uncovering new families and folds in the natural protein universe. Nature, 622(7983), 646–653. https://doi.org/10.1038/s41586-023-06622-3
10. ELIXIR Europe. (2025, January 14-16). AlphaFold Education Summit. https://tess.elixir-europe.org/events/alphafold-education-summit
11. EMBL-EBI. (n.d.). AlphaFold - A practical guide. https://www.ebi.ac.uk/training/online/courses/alphafold/
12. Google DeepMind. (2024, November 11). Google DeepMind and Isomorphic Labs introduce AlphaFold 3, a breakthrough AI model for biology. Google Blog. https://blog.google/technology/ai/google-deepmind-isomorphic-alphafold-3-ai-model/#alphafold-server
13. Guo, S., Meng, Y., Lin, L., Zhou, Z., Li, H., Tian, X., & Huang, W. (2024b). Artificial intelligence alphafold model for molecular biology and drug discovery: a machine-learning-driven informatics investigation. Molecular Cancer, 23(1). https://doi.org/10.1186/s12943-024-02140-6
14. Kasanmascheff, M. (2024). Google DeepMind Open-Sources AlphaFold 3 to Researchers. WinBuzzer. https://winbuzzer.com/2024/11/13/google-deepmind-deepmind-open-sources-alphafold-3-to-researchers-xcxwbn/
15. Krishnaswamy, S. (2024). A cautionary tale for open science: AlphaFold3. People’s Democracy. https://peoplesdemocracy.in/2024/0602_pd/cautionary-tale-open-science-alphafold3
16. McKay, C. (2024, November 11). Google DeepMind makes AlphaFold 3 open source for Academia. Maginative. https://www.maginative.com/article/google-deepmind-makes-alphafold-3-open-source-for-academia/
17. Mirdita, M., Schütze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S., & Steinegger, M. (2022). ColabFold: making protein folding accessible to all. Nature Methods, 19(6), 679–682. https://doi.org/10.1038/s41592-022-01488-1
18. Nji, E., Cramer, K. C., Rüffin, N. V., Fofana, F. G., Heiba, W., & Sankhe, S. (2025). Leveraging AlphaFold for innovation and sustainable health research in Africa. Nature Communications, 16(1). https://doi.org/10.1038/s41467-025-56545-y
19. OMICS Bangladesh. (n.d.). https://omicsbd.com/courses/Mastering-AlphaFold-3
20. Tunyasuvunakool, K., Adler, J., Wu, Z., Green, T., Zielinski, M., Žídek, A., Bridgland, A., Cowie, A., Meyer, C., Agrawal, A., Bates, R., Jain, R., Petersen, S., Clancy, E., Kohli, P., Hassabis, D., & Jumper, J. (2021). Highly accurate protein structure prediction for the human proteome. Nature, 596(7873), 590–596. https://doi.org/10.1038/s41586-021-03828-1
21. United Nations Development Programme. (2025). Human Development Report 2025: A matter of choice people and possibilities in the age of AI. In https://hdr.undp.org/system/files/documents/global-report-document/hdr2025reporten.pdf (No. 9789211542639). RR Donnelly. https://hdr.undp.org/system/files/documents/global-report-document/hdr2025reporten.pdf
22. Varadi, M., & Velankar, S. (2022). The impact of AlphaFold Protein Structure Database on the fields of life sciences. PROTEOMICS, 23(17). https://doi.org/10.1002/pmic.202200128
23. Varadi, M., Anyango, S., Deshpande, M., Nair, S., Natassia, C., Yordanova, G., Yuan, D., Stroe, O., Wood, G., Laydon, A., Žídek, A., Green, T., Tunyasuvunakool, K., Petersen, S., Jumper, J., Clancy, E., Green, R., Vora, A., Lutfi, M., . . . Velankar, S. (2021). AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Research, 50(D1), D439–D444. https://doi.org/10.1093/nar/gkab1061
24. Varadi, M., Bertoni, D., Magana, P., Paramval, U., Pidruchna, I., Radhakrishnan, M., Tsenkov, M., Nair, S., Mirdita, M., Yeo, J., Kovalevskiy, O., Tunyasuvunakool, K., Laydon, A., Žídek, A., Tomlinson, H., Hariharan, D., Abrahamson, J., Green, T., Jumper, J., . . . Velankar, S. (2023). AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Research, 52(D1), D368–D375. https://doi.org/10.1093/nar/gkad1011