Noman, A. | Search bioRxiv

bioRxiv2026

In silico transcriptomic analysis reveals shared molecular signatures and immune-associated pathways between Hashimotos thyroiditis and type 2 diabetes with exploratory drug repurposing

The management of Hashimotos thyroiditis (HT), one of the most prevalent autoimmune disorders worldwide, becomes more complex when it coexists with type 2 diabetes (T2D) compared with the management of either disease alone. This complexity may arise from overlapping genetic, metabolic, and immune dysregulation, as well as potential therapeutic conflicts. Although HT and T2D are known to co-occur and share immune and metabolic features, the molecular characteristics underlying these overlaps have not been systematically explored. This study aimed to identify shared gene expression signatures and associated biological pathways between HT and T2D using an in silico, hypothesis-generating approach, and to explore candidate compounds that may be relevant to both conditions. Independent transcriptomic datasets (GSE138198 for control/HT and GSE29231 for control/T2D) were analyzed, leading to the identification of 59 genes that were differentially expressed in both HT and T2D compared with control samples. Protein-protein interaction (PPI) network analysis prioritized five shared key genes (sKGs): CDC42, CD74, FOS, RAC2, and YWHAB. Functional enrichment analysis of these sKGs revealed overlapping biological processes, molecular functions, cellular components, and immune-related signaling pathways, as well as shared regulatory networks involving transcription factors (FOXC1 and HNF4A) and microRNAs (hsa-miR-221-3p and hsa-miR-29a-3p). Immune infiltration analysis demonstrated broadly similar patterns of immune dysregulation in both diseases, providing additional biological context for the observed shared molecular signatures. Finally, an exploratory in silico drug repurposing pipeline incorporating molecular docking, ADMET profiling, drug-likeness assessment, and molecular dynamics simulations prioritized three candidate compounds: gliquidone, oleanolic acid, and glipizide for further investigation. Overall, this study provides a hypothesis-generating framework highlighting shared molecular features between HT and T2D, which may inform future experimental validation and clinical research. Author SummaryIn this work, we wanted to better understand why Hashimotos thyroiditis, an autoimmune condition that affects the thyroid gland, is often seen in people who also have type 2 diabetes. Treating patients who live with both conditions can be difficult, and we were interested in finding out whether they share common biological causes. To do this, we examined genetic data from individuals with each disease and looked for patterns that appeared in both groups. We discovered several genes that seem to act in similar ways in the two conditions, particularly genes linked to immune system activity and associated pathways. This finding suggests that shared molecular signatures and immune-associated pathways may play a role in the development of both diseases. We also explored how these shared genetic features influence larger biological processes and immune responses. The similarities we found support the idea that the two diseases may be connected through related biological pathways. In addition, we used computer-based screening methods to identify existing drugs that might influence these shared pathways. While these results need further testing, we hope our findings help open new directions for research and eventually contribute to better care for patients affected by both conditions.

bioinformatics↗

bioRxiv2024

Synthetic Histology Images for Training AI Models: A Novel Approach to Improve Prostate Cancer Diagnosis

Prostate cancer (PCa) poses significant challenges for timely diagnosis and prognosis, leading to high mortality rates and increased disease risk and treatment costs. Recent advancements in machine learning and digital imagery offer promising potential for developing automated and objective assessment pipelines that can reduce human capital and resource costs. However, the reliance of AI models on large amounts of clinical data for training presents a significant challenge, as this data is often biased, lacking diversity, and not readily available. Here we aim to address this limitation by employing customized generative adversarial network (GAN) models to produce high-quality synthetic images of different PCa grades (radical prostatectomy (RP)) and needle biopsies, which were customized to account for the granularity associated with each Gleason grade. The generated images were subjected to multiple rounds of benchmarking, quantifications and quality control assessment before being used to train an AI model (EfficientNet) for grading digital histology images of adenocarcinoma specimens (RP sections) and needle biopsies obtained from the PANDA challenge repository. Validation was performed using the AI model trained with synthetic data to grade digital histology from the cancer genome atlas (TCGA) (RP sections) and needle biopsy data from Radboud University Medical Center and Karolinska Institute. Results demonstrated that the AI model trained with a combination of image patches derived from original and enhanced synthetic images outperformed the model trained with original digital histology images. Together, this study demonstrates the potential of customized GAN models to generate a large cohort of synthetic data that can train AI models to effectively grade PCa specimens. This approach could potentially eliminate the need for extensive clinical data for training any AI model in the domain of digital imagery, leading to cost and time-effective diagnosis and prognosis.

cancer biology↗