Wednesday, June 28, 2023

PETs Prize Challenge Track B, Dataset use

 Hello PETs Prize Challenge Track B participants,


Thank you for your participation in the challenge! In case you missed it, the winners were announced last week at the Summit for Democracy. You can learn more about the winners in the U.S. winners blog post or the U.K. winners blog post.


The pandemic forecasting dataset from the challenge has been released online with an accompanying paper. You can find all of that on the dataset website, including instructions for how to cite the dataset. 


You are welcome to share any work that you produced as part of the challenge. Please be sure to acknowledge the challenge and appropriately cite the challenge dataset. You should link to the official release of the dataset rather than making any data available yourselves. 


If you also participated in Track A: Financial Crime Prevention of the challenge, please note that guidelines for sharing Track A work are more expansive and have been sent out separately to Track A participants. 


If you have any further questions, please send them to NIST at PrivacyEng@nist.gov.


Thanks again for your engagement with the challenge. We look forward to following your continued work in privacy-enhancing technologies research. 


Best,

The DrivenData Team and the PETs Prize Challenge organizers


PH AWS quotes

 I have calculated the price for instances suitable for our project.

Based on the information in the following link:

AWS EC2 Instance Types  (section of Accelerated Computing)

I have created an Excel spreadsheet to provide you with a detailed breakdown and comparison. You can access it using the following link:

 

The following link contains PDF files of all the estimations, which I downloaded from AWS:
AWS - PDF

 

Sunday, June 25, 2023

DELTA New Department Leaders Institute

 

I found information about the DELTA New Department Leaders Institute, which is designed to provide new department chairs with the knowledge and tools needed to positively launch and navigate their leadership role. The program is an intermediate-level, instructor-led program that takes place over two days for a total of 12 hours of instruction. The program offers opportunities for participants to interact with facilitators and each other, with ample time dedicated to breakout activities and discussion².


Is this what you were looking for?


Source: Conversation with Bing, 6/25/2023

(1) DELTA New Department Leaders Institute – ASEE Learning. https://learning.asee.org/course_catalog/delta-new-department-leaders-institute/.

(2) Delta Leadership Program 2023 - Delta Protection Commission. https://delta.ca.gov/delta-leadership-program/.

(3) Leader Bios | Delta News Hub. https://news.delta.com/leader-bios.

Saturday, June 24, 2023

food desert and health disparity

Food deserts are indicators of more than just socioeconomic injustice; they indicate public health and safety concerns for those living within their borders. Residents with a chronic lack of access to adequate food resources are shown to have higher rates of diabetes, obesity, and cardiovascular disease¹. Restricted access to healthy foods may magnify health disparities that lead to higher rates of diabetes, hypertension, and cardiovascular disease⁴.

To research how food deserts affect health disparities, you can look at studies that examine the relationship between food insecurity and chronic health conditions such as diabetes, obesity, heart disease, mental health disorders, and other chronic diseases². You can also look at studies that examine the impact of food deserts on different populations, such as low-income communities and communities of color².

(1) https://sites.duke.edu/lit290s-1_02_s2017/2017/03/04/health-and-socioeconomic-disparities-of-food-deserts/#:~:text=Food%20deserts%20are%20indicators%20of%20more%20than%20just,rates%20of%20diabetes%2C%20obesity%2C%20and%20cardiovascular%20disease%20%28Corapi%29.. https://sites.duke.edu/lit290s-1_02_s2017/2017/03/04/health-and-socioeconomic-disparities-of-food-deserts/.

(2) Along With Health Care Disparity, Effects Of Food Deserts Likely Making .... https://blockclubchicago.org/2020/04/10/along-with-health-care-disparity-and-poverty-effects-of-food-deserts-likely-making-covid-19-more-deadly-for-black-chicagoans/.

(3) Food Accessibility, Insecurity and Health Outcomes - NIMHD. https://www.nimhd.nih.gov/resources/understanding-health-disparities/food-accessibility-insecurity-and-health-outcomes.html.

(4) The changing landscape of food deserts - PMC - National Center for .... https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7299236/.

(5) Food insecurity in the US: An explainer and research roundup. https://journalistsresource.org/home/food-insecurity-health/.

(6) Food Deserts: Causes, Effects, and Solutions - Healthline. https://www.healthline.com/nutrition/food-deserts.


Friday, June 9, 2023

AGE

 promoter methylation is associate social factor in dogs


Yuan 2011 longevity between mouse and human Framingham study. 


George Williams, evolutionary theory of aging. antagnistic pleotropic genes. 


Hongjie

Lu Wang, UT San Antonio

Alex UW Seattle


=>single cell aging: 

worm

https://c.elegans.aging.atlas.research.calicolabs.com/

fly 

=> Hevolution

GPT for few-shot genomics analysis across species

=> women in Japan and immigrant in US have different breast-cancer risks, suggesting environmental factors on age-dependent risks, city of hope presentation. 

=> 

Karen Guerrero Vazquez

Authors: Karen Guerrero Vazquez, Pilib Ó Broin, Katarzyna Goljanek-Whysall

Selection of miRNA candidates for the treatment of sarcopenia using network-based analysis and differential expression scoring. Karen Guerrero Vazquez 1, Pilib Ó Broin 1, Katarzyna Goljanek-Whysall 2. 1 School of Mathematical & Statistical Sciences, National University of Ireland Galway, 2 School of Medicine, National University of Ireland Galway. Sarcopenia is a natural consequence of aging and leads to progressive muscle wasting. Currently, there is no cure for this condition, and target identification and validation remain pressing challenges. Many potential therapeutic targets have failed in clinical trials or shown poor association with the disease. This project aims to address these challenges by creating a model of microRNA:target interactions for more efficient in silico selection of potential therapeutic targets for sarcopenia. We conducted a novel network-based analysis of microRNA involvement in aging using RNAseq and microarray data from five studies, with a total of 246 samples of skeletal muscle from healthy participants with ages ranging from 19 to 85 years old. We analyzed young, middle age and older adults and determined interactions between genes coming from co-expression, colocalization, genetic interaction, physical interaction, predicted and shared protein domain calculated with GeneMania. Next, we predicted microRNAs that are putative regulators of shortlisted genes using target prediction tools, such as TargetScan, mirDB and mirTarbase. Finally, we add tissue expression from Diana-miTed and miRNATissue Atlas2. After identifying the network of gene:gene and microRNA:gene interactions, the relevance of each node was calculated using multiple scoring metrics and measures of centrality in order to identify the most likely key regulatory microRNAs and their targets during aging. Our model of microRNA-target interactions is specifically tailored to the context of muscle aging and offers a more comprehensive and accurate representation of the complex regulatory mechanisms involved in muscle aging than existing tools, as it integrates multiple layers of biological information. By identifying a few tens of microRNAs and genes with potential therapeutic power, our model offers a valuable and efficient approach to target identification and validation for the treatment of sarcopenia. This abstract has emanated from research supported in part by a research grant from Science Foundation Ireland under Grant number [18/CRT/6214]

=> 

Explaining the asynchrony of aging through cell population dynamicsMing Yang1, Benjamin R. Harrison1, Daniel E.L. Promislow1,21 Department of Laboratory Medicine and Pathology, University of Washington School of Medicine, Seattle, WA 98195 USA 2 Department of Biology, University of Washington, Seattle, WA 98195 USA Different tissues age at different rates within a single individual. Such asynchrony in aging has been widely observed at multiple levels, from functional hallmarks, such as anatomical structures and physiological processes, to molecular endophenotypes, such as the transcriptome and metabolome. However, we lack a conceptual framework to understand why some components age faster than others. Just as demographic models explain why aging evolves, here we test the hypothesis that demographic differences among cell types, determined by cell-specific differences in turnover rate, can explain why the transcriptome shows signs of aging in some cell types but not others. Through analysis of mouse single-cell transcriptome data across diverse tissues and ages, we find that cell lifespan explains a large proportion of the variation in the age-related increase in transcriptome variance. We further show that long-lived cells are characterized by relatively high expression of genes associated with proteostasis, and that the transcriptome of long-lived cells shows greater evolutionary constraint than short-lived cells. In contrast, in short-lived cell types the transcriptome is enriched for genes associated with DNA repair. Based on these observations, we develop a novel heuristic model that explains how and why aging rates differ among cell types.This work was supported by NIH R01 AG063371 (to D. Promislow and S. Pletcher) and Norn Group Impetus Grant (to D. Promislow).

https://www.biorxiv.org/content/10.1101/2023.05.31.543091v1


fair AI high school camp

 

https://www.bloomberg.com/graphics/2023-generative-ai-bias/

Thursday, June 1, 2023

SHAP tutorial from BING

 

There are several popular tutorials on using SHAP for ML. Some of them include:


1. "SHAP: How to Interpret Machine Learning Models With Python" by Dario Radečić on Better Data Science ¹.

2. "An introduction to explainable AI with Shapley values" on the SHAP documentation website ².

3. "Interpretable Machine Learning using SHAP — theory and applications" by Towards Data Science ³.

4. "Tutorial: Explainable Machine Learning with Python and SHAP" by Natalie Beyer on ML Conference ⁴.


These tutorials provide a step-by-step guide on how to use the SHAP package in Python to interpret machine learning models.


Is there anything else you would like to know?


Source: Conversation with Bing, 6/1/2023

(1) SHAP: How to Interpret Machine Learning Models With Python. https://betterdatascience.com/shap/ .

(2) An introduction to explainable AI with Shapley values — SHAP latest .... https://shap.readthedocs.io/en/latest/example_notebooks/overviews/An%20introduction%20to%20explainable%20AI%20with%20Shapley%20values.html.

(3) Interpretable Machine Learning using SHAP — theory and applications .... https://towardsdatascience.com/interpretable-machine-learning-using-shap-theory-and-applications-26c12f7a7f1a.

(4) Tutorial: Explainable Machine Learning with Python and SHAP. https://mlconference.ai/blog/tutorial-explainable-machine-learning-with-python-and-shap/.

SHAP tutorial

 

KernelSHAP vs TreeSHAP

Comparing SHAP approximation methods based on speed, complexity and other considerations

medium-articles/kernelSHAP_vs_treeSHAP.ipynb at master · conorosully/medium-articles · GitHub