Biology subjects

Shneyderman, A.

Publications and source records attributed to Shneyderman, A..

3 recordsLinked to original sources

bioRxiv2024

Comparative analysis of Endoxifen, Tamoxifen and Fulvestrant: A Bioinformatics Approach to Uncover Mechanisms of Action in Breast Cancer

Breast cancer remains a significant health challenge, with estrogen receptor positive (ER+) subtypes being particularly prevalent forms of breast cancer. Current anti-estrogen therapies, such as tamoxifen and fulvestrant, have limitations, including partial agonist activity and resistance development, which evidence the need for more potent alternatives. Endoxifen, a metabolite of tamoxifen, has emerged as a promising breast cancer therapeutic candidate due to its superior anti-estrogenic effects and side effect profile. The omics signatures for endoxifen, tamoxifen and fulvestrant, obtained from publicly available datasets, were aggregated and harmonized by means of the PandaOmics platform, a commercially available target-discovery platform using multiple AI engines including generative pretrained transformers. Pathway enrichment analyses provided insight into these agents mechanisms of action (MOA) in breast cancer. The analyses revealed unexpected variances in several key pathways from expected interactions via estrogen-dependent and independent effects. All three drugs downregulated estrogen signaling and cell cycle-related pathways, such as E2F targets, G2-M checkpoints, Myc targets, and mitotic spindle, and stimulated apoptosis. Fulvestrant and tamoxifen activated pro-inflammatory and immune pathways and perturbed epithelial-mesenchymal transition (EMT). Endoxifen perturbed the PI3K/Akt/mTORC1 pathway, pursuant to distinct molecular mechanisms compared to its parent compound, tamoxifen, and fulvestrant. In summary, advanced AI-driven methodologies demonstrate the capacity to analyze multi-omics data in a comparative way to advance the understanding of endocrine therapy mechanisms in breast cancer. This insight into the distinct effects of endoxifen, tamoxifen, and fulvestrant may aid in selecting the most effective therapies for specific indications and in identifying drug-specific biomarkers.

cancer biology↗

bioRxiv2024

Precious3GPT: Multimodal Multi-Species Multi-Omics Multi-Tissue Transformer for Aging Research and Drug Discovery

We present a multimodal multi-species multi-omics multi-tissue transformer for aging research and drug discovery capable of performing multiple tasks such as age prediction across species, target discovery, tissue, sex, and disease sample classification, drug sensitivity prediction, replication of omics response and prediction of biological and phenotypic response to compound treatment. This model combines textual, tabular, and knowledge graph-derived representations of biological experiments to provide insights into molecular-level biological processes. We demonstrate that P3GPT has developed an intuition for the interactions between compounds, pathologies, and gene regulation in the context of multiple species and tissues. In these areas, it outperforms existing LLMs and we highlight its utility in diverse case studies. P3GPT is a general model that may be used as a target identification tool, aging clock, digital laboratory, and scientific assistant. The model is intended as a community resource available open source as well as via a Discord server.

bioinformatics↗

bioRxiv2023

Applying Artificial Intelligence to Identify Common Targets for Treatment of Asthma, Eczema, and Food Allergy

Allergic disorders are common diseases marked by the abnormal immune response towards foreign antigens that are not pathogens. Often patients with food allergy also suffer from asthma and eczema. Given the similarities of these diseases and a shortage of effective treatments, developing novel therapeutics against common targets of multiple allergies would offer an efficient and cost-effective treatment for patients. Herein, we employed the artificial intelligence-driven target discovery platform, PandaOmics, to identify common targets for treating asthma, eczema, and food allergy. Thirty-two case-control comparisons were generated from 15, 11, and 6 transcriptomics datasets related to asthma (558 cases, 315 controls), eczema (441 cases, 371 controls), and food allergy (208 cases, 106 controls) respectively, and allocated into three meta-analyses for target identification. Top-100 high-confidence targets and Top-100 novel targets were prioritized by PandaOmics for each allergic disease. Six common high-confidence targets (i.e., IL4R, IL5, JAK1, JAK2, JAK3, and NR3C1) across all three allergic diseases have approved drugs for treating asthma and eczema. Based on the targets dysregulated expression profiles and their mechanism of action in allergic diseases, three potential therapeutic targets were proposed. IL5 was selected as a high-confidence target due to its strong involvement in allergies. PTAFR was identified for drug repurposing, while RNF19B was selected as a novel target for therapeutic innovation. Analysis of the dysregulated pathways commonly identified across asthma, eczema, and food allergy revealed the well-characterized disease signature and novel biological processes that may underlie the pathophysiology of allergies. Altogether, our study dissects the shared pathophysiology of allergic disorders and reveals the power of artificial intelligence in the exploration of novel therapeutic targets.

immunology↗