Search results for: Babak Rezaei

Authors: Babak Forouraghi

Abstract:

A genetic circuit is a collection of interacting genes and proteins that enable individual cells to implement and perform vital biological functions such as cell division, growth, death, and signaling. In cell engineering, synthetic gene circuits are engineered networks of genes specifically designed to implement functionalities that are not evolved by nature. These engineered networks enable scientists to tackle complex problems such as engineering cells to produce therapeutics within the patient's body, altering T cells to target cancer-related antigens for treatment, improving antibody production using engineered cells, tissue engineering, and production of genetically modified plants and livestock. Construction of computational models to realize genetic circuits is an especially challenging task since it requires the discovery of flow of genetic information in complex biological systems. Building synthetic biological models is also a time-consuming process with relatively low prediction accuracy for highly complex genetic circuits. The primary goal of this study was to investigate the utility of a pre-trained bidirectional encoder transformer that can accurately predict gene expressions in genetic circuit designs. The main reason behind using transformers is their innate ability (attention mechanism) to take account of the semantic context present in long DNA chains that are heavily dependent on spatial representation of their constituent genes. Previous approaches to gene circuit design, such as CNN and RNN architectures, are unable to capture semantic dependencies in long contexts as required in most real-world applications of synthetic biology. For instance, RNN models (LSTM, GRU), although able to learn long-term dependencies, greatly suffer from vanishing gradient and low-efficiency problem when they sequentially process past states and compresses contextual information into a bottleneck with long input sequences. In other words, these architectures are not equipped with the necessary attention mechanisms to follow a long chain of genes with thousands of tokens. To address the above-mentioned limitations of previous approaches, a transformer model was built in this work as a variation to the existing DNA Bidirectional Encoder Representations from Transformers (DNABERT) model. It is shown that the proposed transformer is capable of capturing contextual information from long input sequences with attention mechanism. In a previous work on genetic circuit design, the traditional approaches to classification and regression, such as Random Forrest, Support Vector Machine, and Artificial Neural Networks, were able to achieve reasonably high R2 accuracy levels of 0.95 to 0.97. However, the transformer model utilized in this work with its attention-based mechanism, was able to achieve a perfect accuracy level of 100%. Further, it is demonstrated that the efficiency of the transformer-based gene expression classifier is not dependent on presence of large amounts of training examples, which may be difficult to compile in many real-world gene circuit designs.

Keywords: transformers, generative ai, gene expression design, classification

Procedia PDF Downloads 29

Authors: Babak Forouraghi

Abstract:

A genetic circuit is a collection of interacting genes and proteins that enable individual cells to implement and perform vital biological functions such as cell division, growth, death, and signaling. In cell engineering, synthetic gene circuits are engineered networks of genes specifically designed to implement functionalities that are not evolved by nature. These engineered networks enable scientists to tackle complex problems such as engineering cells to produce therapeutics within the patient's body, altering T cells to target cancer-related antigens for treatment, improving antibody production using engineered cells, tissue engineering, and production of genetically modified plants and livestock. Construction of computational models to realize genetic circuits is an especially challenging task since it requires the discovery of the flow of genetic information in complex biological systems. Building synthetic biological models is also a time-consuming process with relatively low prediction accuracy for highly complex genetic circuits. The primary goal of this study was to investigate the utility of a pre-trained bidirectional encoder transformer that can accurately predict gene expressions in genetic circuit designs. The main reason behind using transformers is their innate ability (attention mechanism) to take account of the semantic context present in long DNA chains that are heavily dependent on the spatial representation of their constituent genes. Previous approaches to gene circuit design, such as CNN and RNN architectures, are unable to capture semantic dependencies in long contexts, as required in most real-world applications of synthetic biology. For instance, RNN models (LSTM, GRU), although able to learn long-term dependencies, greatly suffer from vanishing gradient and low-efficiency problem when they sequentially process past states and compresses contextual information into a bottleneck with long input sequences. In other words, these architectures are not equipped with the necessary attention mechanisms to follow a long chain of genes with thousands of tokens. To address the above-mentioned limitations, a transformer model was built in this work as a variation to the existing DNA Bidirectional Encoder Representations from Transformers (DNABERT) model. It is shown that the proposed transformer is capable of capturing contextual information from long input sequences with an attention mechanism. In previous works on genetic circuit design, the traditional approaches to classification and regression, such as Random Forrest, Support Vector Machine, and Artificial Neural Networks, were able to achieve reasonably high R2 accuracy levels of 0.95 to 0.97. However, the transformer model utilized in this work, with its attention-based mechanism, was able to achieve a perfect accuracy level of 100%. Further, it is demonstrated that the efficiency of the transformer-based gene expression classifier is not dependent on the presence of large amounts of training examples, which may be difficult to compile in many real-world gene circuit designs.

Keywords: machine learning, classification and regression, gene circuit design, bidirectional transformers

Procedia PDF Downloads 33

Authors: Mohammad H. Yarmohammadian, Fatemeh Rezaei

Abstract:

Introduction: Improving patient safety in health systems is one of the main priorities in healthcare systems, so clinical risk management in organizations has become increasingly significant. Although several tools have been developed for clinical risk management, each has its own limitations. Aims: This study aims to develop a comprehensive tool that can complete the limitations of each risk assessment and management tools with the advantage of other tools. Methods: Procedure was determined in two main stages included development of an initial model during meetings with the professors and literature review, then implementation and verification of final model. Subjects and Methods: This study is a quantitative − qualitative research. In terms of qualitative dimension, method of focus groups with inductive approach is used. To evaluate the results of the qualitative study, quantitative assessment of the two parts of the fourth phase and seven phases of the research was conducted. Purposive and stratification sampling of various responsible teams for the selected process was conducted in the operating room. Final model verified in eight phases through application of activity breakdown structure, failure mode and effects analysis (FMEA), healthcare risk priority number (RPN), root cause analysis (RCA), FT, and Eindhoven Classification model (ECM) tools. This model has been conducted typically on patients admitted in a day-clinic ward of a public hospital for surgery in October 2012 to June. Statistical Analysis Used: Qualitative data analysis was done through content analysis and quantitative analysis done through checklist and edited RPN tables. Results: After verification the final model in eight-step, patient's admission process for surgery was developed by focus discussion group (FDG) members in five main phases. Then with adopted methodology of FMEA, 85 failure modes along with its causes, effects, and preventive capabilities was set in the tables. Developed tables to calculate RPN index contain three criteria for severity, two criteria for probability, and two criteria for preventability. Tree failure modes were above determined significant risk limitation (RPN > 250). After a 3-month period, patient's misidentification incidents were the most frequent reported events. Each RPN criterion of misidentification events compared and found that various RPN number for tree misidentification reported events could be determine against predicted score in previous phase. Identified root causes through fault tree categorized with ECM. Wrong side surgery event was selected by focus discussion group to purpose improvement action. The most important causes were lack of planning for number and priority of surgical procedures. After prioritization of the suggested interventions, computerized registration system in health information system (HIS) was adopted to prepare the action plan in the final phase. Conclusion: Complexity of health care industry requires risk managers to have a multifaceted vision. Therefore, applying only one of retrospective or prospective tools for risk management does not work and each organization must provide conditions for potential application of these methods in its organization. The results of this study showed that the integrated clinical risk management model can be used in hospitals as an efficient tool in order to improve clinical governance.

Keywords: failure modes and effective analysis, risk management, root cause analysis, model

Procedia PDF Downloads 219