Background: Specific subvariants of uveal melanoma (UM) are associated with increased rates of metastasis compared to other subvariants. BRCA1 (BReast CAncer gene 1)-associated protein-1 (BAP1) is encoded by a gene that has been linked to aggressive behavior in UM. Methods: We evaluated BAP1 for the presence of intrinsically disordered protein regions (IDPRs) and its protein–protein interactions (PPI). We evaluated specific sequence-based features of the BAP1 protein using a set of bioinformatic databases, predictors, and algorithms. Results: We show that BAP1’s structure contains extensive IDPRs as it is highly enriched in proline residues (the most disordered amino acid; p-value < 0.05), the average percent of predicted disordered residues (PPDR) was 57.34%, and contains 9 disorder-based binding sites (ie. molecular recognition features (MoRFs)). BAP1’s intrinsic disorder allows it to engage in a complex PPI network with at least 49 partners (p-value < 1.0 × 10⁻¹⁶). Conclusion: These findings show that BAP1 contains IDPRs and an intricate PPI network. Mutations in UM that are associated with the BAP1 gene may alter the function of the IDPRs embedded into its structure. These findings develop the understanding of UM and may provide a target for potential novel therapies to treat this aggressive neoplasm.

In recent years, there has been tremendous enthusiasm with respect to detailing the genetic basis of many neoplasms, including conjunctival melanoma (CM). We aim to analyze five proteins associated with CM, namely BRAF, NRAS, c-KIT, NF1, and PTEN. We evaluated each protein for its intrinsically disordered protein regions (IDPRs) and its protein-protein interactions (PPI) with the Predictor of Natural Disordered Protein Regions (PONDR^®) and the Search Tool for the Retrieval of Interacting Genes (STRING^®). Our PONDR^® analysis found high levels of IDPRs in all five proteins with mutations linked to CM. The highest levels of IDPRs were in BRAF (45.95%), followed by PTEN (31.76%), NF1 (22.19%), c-KIT (21.82%), and NRAS (14.81%). Our STRING analysis found that each of these five proteins had more predicted interactions then expected (p-value < 1.0 × 10⁻¹⁶). Our analysis demonstrates that the mutations linked to CM likely affected IDPRs and possibly altered their highly complex PPIs. Quantifying IDPRs in BRAF, NRAS, c-KIT, NF1, and PTEN and understanding these protein regions are important processes as IDPRs can be possible drug targets for novel targeted therapies for treating CM.

• Our study demonstrated that intrinsic disorder is abundant in the aqueous humor. • The 749 aqueous proteins analyzed were enriched with disorder-promoting residues. • 208 aqueous humor proteins were predicted to be highly intrinsically disordered. • Misregulation of IDPs may promote pathology in the aqueous humor. • IDPs in aqueous humor may serve as future targets for novel therapeutics.

BACKGROUND: Artificial Intelligence (AI) models have shown potential in various educational contexts. However, their utility in explaining complex biological phenomena, such as Intrinsically Disordered Proteins (IDPs), requires further exploration. This study empirically evaluated the performance of various Large Language Models (LLMs) in the educational domain of IDPs. METHODS: Four LLMs, GPT-3.5, GPT-4, GPT-4 with Browsing, and Google Bard (PaLM 2), were assessed using a set of IDP-related questions. An expert evaluated their responses across five categories: accuracy, relevance, depth of understanding, clarity, and overall quality. Descriptive statistics, ANOVA, and Tukey's honesty significant difference tests were utilized for analysis. RESULTS: The GPT-4 model consistently outperformed the others across all evaluation categories. Although GPT-4 and GPT-3.5 were not statistically significantly different in performance (p>0.05), GPT-4 was preferred as the best response in 13 out of 15 instances. The AI models with browsing capabilities, GPT-4 with Browsing and Google Bard (PaLM 2) displayed lower performance metrics across the board with statistically significant differences (p<0.0001). CONCLUSION: Our findings underscore the potential of AI models, particularly LLMs such as GPT-4, in enhancing scientific education, especially in complex domains such as IDPs. Continued innovation and collaboration among AI developers, educators, and researchers are essential to fully harness the potential of AI for enriching scientific education.