ODU Proposal Development support.

 ODU Proposal Development support. 

Below is a link to ODU’s Proposal Development support. If you click the Request Grant Proposal Support link on the bottom, you can request assistance from a Grant Development Specialist. Their services are optional, but can be a real value add to a proposal!

 

https://ww1.odu.edu/facultystaff/research/development/proposal-development

 

ODU Sponsored Programs

Old Dominion University Research Foundation

4111 Monarch Way, Suite 204

Norfolk, VA 23508

Proposal Development

From an original idea to actual submission, creating and writing a project proposal involves an intricate process requiring perseverance, organizational and time management skills; interpersonal negotiation and communication skills; the ability to accept feedback, often in the form of criticism; and the ability to handle rejection, make corrections and adjustments, and submit again.

The Grant Development Specialists in the Office of Research are available to provide feedback and editorial assistance on the non-technical aspects of your proposal by ensuring that you have adequately responded to all of the requirements of the Request for Proposal (RFP); by making sure that you have identified the University and its various colleges, departments and other features appropriately; and by providing feedback regarding the readability, logic and flow of the narrative.

The links below will help you through the process by providing resources to produce a grammatically correct document that is free of spelling errors, tips and pointers for writing proposals whether they are to agencies or to foundations, and ending with information on how to design and implement an evaluation component for your project.

• REQUEST GRANT PROPOSAL SUPPORT

AI Institute virtual organization

 

https://aiinstitutes.org/

AWS AI in healthcare

 

https://www.amazon.science/news-and-features/how-amazon-and-columbia-university-are-collaborating-to-advance-ai-in-healthcare

Sunita Mishra, chief medical officer at Amazon Health Services, and Katrina Armstrong, dean of Columbia's Faculties of Health Sciences and the Vagelos College of Physicians and Surgeons, recently shared their thoughts on how AI could change healthcare for the better.

DL and GWAS

 

 limimitation of DL vs GWAS. DL can learn to use wildtype feature for classification. 

 

discuss deepGWAS

mobilize data

 

The Mobilize Center brings together individuals from diverse fields to advance computational methods for utilizing novel data sources in movement research

https://mobilize.stanford.edu/data/available-datasets/

MD trajectory data base

 

https://simtk.org/frs/index.php?group_id=285

Eric Volz, sars-cov-2 logistic growth

Volz 2023 show logistic growth does not fit well with Omicro data 

 https://www.nature.com/articles/s41576-023-00610-z

GPT: 

The key findings of Erik Volz's paper, "Fitness, Growth and Transmissibility of SARS-CoV-2 Genetic Variants," published in Nature Reviews Genetics, 2023, include:

1. Genomic Surveillance and Variant Detection:

   • The massive global sequencing effort for SARS-CoV-2 has enabled rapid detection and assessment of new variants. This has led to the development of new strategies for characterizing the fitness and transmissibility of emerging variants, utilizing both classic population genetics models and contemporary epidemiological and phylodynamic analyses.

2. Evolution and Adaptation of SARS-CoV-2:

   • The COVID-19 pandemic saw multiple waves driven by distinct SARS-CoV-2 lineages, each with significant changes in transmissibility, antigenicity, and pathogenicity. These variants, such as Alpha, Beta, and Gamma, were identified through genomic surveillance before increases in infections and hospitalizations were evident in epidemiological data.

3. Fitness and Selection Coefficient:

   • The concept of fitness in population genetics, particularly the selection coefficient, is crucial for understanding how quickly advantageous variants become dominant. The logistic growth model, despite its simplicity, has been applied to estimate the selection coefficient for various SARS-CoV-2 variants, providing a fast means to measure growth across many variants.

4. Complexity in Variant Growth:

   • Variant growth is influenced by factors beyond differential fitness, such as genetic drift, founder effects, and epidemiological or behavioral factors. The paper highlights the challenges in disentangling these effects to accurately measure variant fitness.

5. Mechanisms of Fitness Advantage:

   • The paper discusses the importance of virological evidence in evaluating the mechanisms behind a variant's fitness advantage, such as non-synonymous changes in the spike protein that influence virus fitness.

6. Consistent Growth Across Regions:

   • Confidence in a variant's fitness advantage increases with consistent growth observed across different regions and demographics. This was exemplified by the growth patterns of the Delta variant, which showed consistent high growth rates across multiple regions and demographic groups.

7. Phylodynamic Analysis:

   • Phylodynamic modeling provides independent validation of epidemiological trends by analyzing genetic diversity and epidemic history. This approach has been used to estimate the growth rates and reproduction numbers of various SARS-CoV-2 lineages.

8. Epidemiological Models and Data Integration:

   • The integration of epidemiological models with genomic data has been essential in estimating variant-specific reproduction numbers and understanding the dynamics of competing variants. These models consider the impact of transmissibility, immune evasion, and generation times on variant growth.

9. Future Perspectives:

   • The paper concludes that as COVID-19 transitions into a manageable infectious disease, genomic surveillance will remain crucial for monitoring and responding to new variants. Methods developed during the SARS-CoV-2 pandemic will be applicable to other pathogens, enhancing future public health responses.

These findings emphasize the importance of genomic surveillance and advanced modeling techniques in understanding and managing the evolution and spread of SARS-CoV-2 and other emerging pathogens​​.

Phylodynamic analysis is a powerful method used to understand the growth and spread of pathogen variants by integrating phylogenetic (genetic) and epidemiological data. Here's how phylodynamic analysis can estimate variant growth trends:

1. Genetic Data Collection and Phylogenetic Tree Construction

• Genetic Sequencing: First, viral genomes from infected individuals are sequenced.
• Phylogenetic Trees: These sequences are then used to construct a phylogenetic tree, which shows the evolutionary relationships between different viral strains. Each branch point (node) represents a common ancestor.

2. Modeling Population Dynamics

• Coalescent Models: These models trace the genealogy of sampled genomes backward in time to estimate the rate at which new infections occur. The shape and branching patterns of the tree reflect the population dynamics of the virus.
• Birth-Death Models: These models estimate the rate of new infections (births) and recoveries or deaths (deaths) over time, providing insights into how quickly the virus spreads and declines.

3. Estimating Growth Rates

• Exponential Growth Detection: During an epidemic's exponential growth phase, the tree will show many closely related sequences branching out rapidly. This abundance of recent, closely related lineages indicates a high growth rate.
• Effective Reproduction Number (Rt): Phylodynamic models can estimate Rt, which is the average number of secondary infections caused by a single infected individual. Changes in Rt over time can indicate how control measures or changes in virus transmissibility impact growth.

4. Assessing Selection Pressure and Variant Fitness

• Mutation Analysis: By examining specific mutations within the phylogenetic tree, researchers can identify which variants are spreading more rapidly. Mutations that frequently appear in branches leading to many descendant sequences suggest a fitness advantage.
• Selective Sweeps: If a new variant with advantageous mutations spreads rapidly, it will dominate the phylogenetic tree, indicating a selective sweep. The speed and extent of this sweep can be quantified to estimate the variant's fitness.

5. Temporal and Geographic Trends

• Temporal Signal: Phylodynamic methods can incorporate the sampling dates of sequences to provide a temporal framework. This allows estimation of when a variant emerged and how its prevalence changed over time.
• Geographic Spread: By combining phylogenetic data with geographic information, researchers can track how variants spread between regions. Clades (branches) of the tree can be associated with different locations, showing the movement and expansion of variants.

6. Integration with Epidemiological Data

• Corroboration with Case Data: Phylodynamic estimates can be validated and refined using epidemiological data, such as case counts, hospitalizations, and deaths. This combined approach provides a comprehensive view of variant dynamics.
• Real-Time Analysis: Continuous genomic surveillance and real-time phylodynamic analysis enable early detection of emerging variants with significant growth advantages, guiding public health responses.

Example: Delta Variant Growth in the UK

In the paper by Volz, phylodynamic analysis was applied to understand the growth of the Delta variant in the UK. The study constructed phylogenetic trees from sampled genomes and used birth-death models to estimate the effective reproduction number of Delta. The consistent high growth rates across different regions and demographics provided robust evidence of Delta's fitness advantage.

Phylodynamic analysis thus offers a robust framework for estimating the growth trends of viral variants by leveraging genetic data to infer epidemiological dynamics, helping to predict and mitigate the impact of emerging pathogens.

CLIMATE CHANGE AND HUMAN HEALTH SEED GRANTS

 CLIMATE CHANGE AND HUMAN HEALTH SEED GRANTS

https://www.bwfund.org/funding-opportunities/climate-change-and-human-health/climate-change-and-human-health-seed-grants/grant-recipients/

The Bloch Sphere (simply explained)

 

two qubit base vectors

 

20240710 UTC quantum circuit

https://www.youtube.com/playlist?list=PLo62Vf9IYqhosCIlWLtV15hb3rUUwAcpX

* IBM, right click to see icon info, gate, NOT, CNOT. Drag and drop on gate. editing the code. 

* explain Hadamard gate

* explain Q-sphere

* explain CNOT

q0 + Hadamard + CNOT = Entangled two-qubit state

https://youtu.be/wCAZWlcS-3I?si=EXaBNr7Wp6_fyuEH

super-position versus entaglement

https://youtu.be/0fGyJbqdCfY?si=3COYTQqolrFZfvIV

https://youtu.be/mObvTSakK5Y?si=3-GUqqwS7nhf3HRx

interference

https://youtu.be/evm38AMeJyk?si=S63eIX9Uu2aSSKk5

1. introduction to IBM quantum circuit. Go over the icon for information. Quits. Q-sphere. 

Entanglement example:

 

https://quantum.ibm.com/composer/files/new

https://github.com/hongqin/quantum_sandbox

 review Bloch sphere definition, using IBM Quantum Composer to interactively explore the states, quantum state definition, 

one qubit. Tip: manually edit the script for fast editing. 

two qubit

Q-sphere view

https://learning.quantum-computing.ibm.com/tutorial/composer-user-guide#q-sphere-view 

Midterm review? 

math for quantum computing

 

math for quantum computing

transformer by hand

 transformer by hand

https://aibyhand.substack.com/p/8-can-you-calculate-a-transformer

https://www.youtube.com/watch?v=v14_yAqY0X0

GenoTEX

GenoTEX: A Benchmark for Evaluating LLM-Based Exploration of Gene Expression Data in Alignment with Bioinformaticians

Haoyang Liu, Haohan Wang

https://arxiv.org/abs/2406.15341 

https://arxiv.org/abs/2402.12391

https://github.com/Liu-Hy/GenoTex

Paving a New Era of AI-Driven Science: Benchmark and Agent Solutions to Automate End-to-End Scientific Discovery

🔭 Imagine AI agents working as scientists, handling every step of research from picking datasets to creating hypotheses, just like bioinformaticians do. These agents change how AI is used, taking on detailed analyses and result interpretation—tasks usually done by human experts.

🚀 Building upon this innovative concept, we are excited to unveil GenoTEX, a transformative benchmark for automating scientific discovery processes in genomics, focusing on gene expression data. GenoTEX is designed not merely as a dataset but as a comprehensive suite to evaluate and enhance AI-driven methods in genomics data analysis. By establishing this benchmark, we aim to pioneer advancements where our AI agents—our GenoAgents—emulate bioinformaticians to perform comprehensive, end-to-end scientific investigations.

🧬 About GenoTEX:
GenoTEX covers the full cycle of scientific research, from selecting and preprocessing data to performing detailed statistical analysis. The benchmark follows a standardized pipeline, checked by expert bioinformaticians, to ensure the AI-generated results are accurate and reliable. Importantly, the datasets in GenoTEX are manually curated, involving 181 relevant datasets with 163 successfully preprocessed, showcasing over 71,669 lines of manually written code for analysis. This benchmark provides extensive annotated code and results for 1,146 gene identification problems—both unconditional (82) and conditional (1,064)—which serve as a robust foundation for developing and evaluating AI methods in genomics.

🤖 Introducing GenoAgents:
Alongside GenoTEX, we introduce GenoAgents, a suite of AI-driven agents crafted to tackle the nuanced tasks traditionally reserved for human experts. These agents are designed with capabilities that include context-aware planning, iterative correction, and domain expertise integration, simulating a collaborative environment similar to a team of human researchers. Our evaluation shows that GenoAgents can automate the process of gene expression data analysis with good overall accuracy, presenting a promising method for future research in genomics.

Our work highlight the potential of LLM-based approaches to significantly reduce the labor-intensive aspects of gene data analysis. We invite the scientific and tech community to explore GenoTEX and participate in evolving this benchmark. Your insights and contributions are crucial as we strive to refine AI applications in real-world scientific contexts.

Benchmark arXiv link: https://lnkd.in/gkZiNehN
previous workshop version: https://lnkd.in/gXzXcEX6
GitHub link: https://lnkd.in/gnrMni3u

nih Advancing Health Research through Ethical, Multimodal AI

 nih Advancing Health Research through Ethical, Multimodal AI

https://datascience.nih.gov/artificial-intelligence/MultimodalAI

https://grants.nih.gov/funding/searchguideNew/index.html#/?query=%22data%20science%22%20(%22artificial%20intelligence%22%20%22big%20data%22%20readiness%20software%20training%20bioinformatics%20fhir%20cloud%20%22AIM-AHEAD%22%20%22data%20ecosystem%22%20%22FAIR%20data%22%20%22common%20data%22%20computational%20%22computational%20biology%22%20%22data%20integration%22%20%22data%20standards%22%20%22data%20repository%22%20%22data%20repositories%22%20%22deep%20learning%22%20diversity%20informatics%20knowledgebase%20%22machine%20learning%22%20%22natural%20language%20processing%22%20%22neural%20networks%22%20%22quantitative%20science%22%20%22quantum%20computing%22%20RADx%20STRIDES%20%22statistical%20modeling%22%20statistics%20workforce)&type=active,activenosis&foa=all&parent_orgs=all&orgs=all&ac=all&ct=all&pfoa=all&date=01021991-03022023&fields=all&spons=true