Biology subjects

Gholami, S.

Publications and source records attributed to Gholami, S..

3 recordsLinked to original sources

bioRxiv2024

EYE-Llama, an in-domain large language model for ophthalmology

1BackgroundTraining Large Language Models (LLMs) with in-domain data can significantly enhance their performance, leading to more accurate and reliable question-answering (Q&A) systems essential for supporting clinical decision-making and educating patients. MethodsThis study introduces ophthalmic LLMs trained on in-domain, well-curated datasets. We present an open-source substantial ophthalmic language dataset for model training. Our models (EYE-Llama), were pre-trained on an ophthalmology-specific dataset, including paper abstracts, textbooks, and Wikipedia articles. Subsequently, the models underwent fine-tuning using a diverse range of QA pairs. Our models were compared to baseline Llama 2, ChatDoctor, Meditron, Llama 3, and ChatGPT (GPT3{middle dot}5) models, using four distinct test sets, and evaluated quantitatively (Accuracy, F1 score, BERTScore, BARTScore and BLEU score) and qualitatively by two ophthalmologists. FindingsUpon evaluating the models using the synthetic dialogue test set with three different metrics (BERTScore, BARTScore, and BLEU score), our models demonstrated superior performance. Specifically, when evaluated using BERTScore, our models surpassed Llama 2, Llama 3, Meditron, and ChatDoctor in terms of F1 score, and performed on par with ChatGPT, which was trained with 175 billion parameters (EYE-Llama: 0.57, Llama 2: 0.56, Llama 3: 0.55, Meditron: 0.50, ChatDoctor: 0.56, and ChatGPT: 0.57). Additionally, the EYE-Llama model outperformed the above models when evaluated using BARTScore and BLEU scores. When tested on the MedMCQA test set, the fine-tuned models exhibited higher accuracy compared to Llama 2, Meditron, and ChatDoctor models (EYE-Llama: 0.39, Llama 2: 0.33, ChatDoctor: 0.29, Meditron: 0.22). However, ChatGPT, and Llama 3 models outperformed EYE-Llama, achieving accuracies of 0.55, 0.78, and 0.90, respectively. On the PubmedQA test set, our model showed improved accuracy over all other models (EYE-Llama: 0.96, Llama 2: 0.90, Llama 3: 0.92, Meditron: 0.76, ChatGPT: 0.93, ChatDoctor: 0.92). InterpretationThe study shows that pre-training and fine-tuning LLMs like EYE-Llama enhances their performance in specific medical domains. Our EYE-Llama models surpass baseline Llama 2 in all evaluations, highlighting the effectiveness of specialized LLMs in medical QA systems. FundingFunded by NEI R15EY035804 (MNA), R21EY035271 (MNA), and UNC Charlotte Faculty Research Grant (MNA)

bioengineering↗

bioRxiv2024

Transport mechanism of DgoT, a bacterial homolog of SLC17 organic anion transporters

The solute carrier 17 (SLC17) family contains anion transporters that accumulate neuro-transmitters in secretory vesicles, remove carboxylated monosaccharides from lysosomes, or extrude organic anions from the kidneys and the liver. We combined classical molecular dynamics simulations, Markov state modeling and hybrid first principles quantum mechani-cal/classical mechanical (QM/MM) simulations with experimental approaches to describe the transport mechanisms of a model bacterial protein, the D-galactonate transporter DgoT, at atomic resolution. We found that protonation of D46 and E133 precedes galactonate binding and that substrate binding induces closure of the extracellular gate, with the conserved R47 coupling substrate binding to transmembrane helix movement. After isomerization to an inward-facing conformation, deprotonation of E133 and subsequent proton transfer from D46 to E133 opens the intracellular gate and permits galactonate dissociation either in its unprotonated form or after proton transfer from E133. After release of the second proton, apo DgoT returns to the outward-facing conformation. Our results provide a framework to understand how various SLC17 transport functions with distinct transport stoichiometries can be attained through subtle variations in proton and substrate binding/unbinding.

biophysics↗

bioRxiv2023

Federated learning for diagnosis of age-related macular degeneration

This paper presents a federated learning (FL) approach to train deep learning models for classifying age-related macular degeneration (AMD) using optical coherence tomography image data. We employ the use of residual networks and vision transformer encoders for the normal vs AMD binary classification, integrating four unique domain adaptation techniques to address domain shift issues caused by heterogeneous data distribution in different institutions. Experimental results indicate that FL strategies can achieve competitive performance similar to centralized models even though each local model has access to a portion of the training data. Notably, Adaptive Personalization FL strategy stood out in our FL evaluations, consistently delivering high performance across all tests due to its additional local model. Furthermore, the study provides valuable insights into the efficacy of simpler architectures in image classification tasks, particularly in scenarios where data privacy and decentralization are critical using both encoders. It suggests future exploration into deeper models and other FL strategies for a more nuanced understanding of these models performance.

bioengineering↗