Researchers at the University of Warwick (Coventry, UK) and Fudan U. (Shanghai, China) have found that abnormal blood plasma levels of four specific proteins out of the 1,463 normally found predicted disease up to 15 years in advance of diagnosis of all forms of dementia. The researchers used AI (artificial intelligence) to analyze plasma proteins from 52,645 normal patients who had blood plasma frozen in 2006 to 2012 and then analyzed 10-15 years later. The AI picked out several proteins that had previously been linked with dementia in small studies to be accurately predictive of dementia.
The AI analysis in this massive study combined two kinds of data, (1) patient demographics (age, sex, education, genetics) with (2) observation of abnormal plasma levels of either of two proteins alone GFAP (glial fibrillary acidic protein) or GDF15 (growth/differentiation factor 15) and found a tight correlation. Based on the data used in the study, the AI analysis was 89.1% accurate in predicting dementia from all causes, 87.2% accurate in predicting Alzheimer's Disease and 91.2% in predicting vascular dementia.
If this research is repeated and confirmed, the results from this research are good enough to use a simple blood test for GFAP and/or GDF15 for widespread routine screening to identify patients highly likely to develop dementia years later. Once this this test for dementia is independently confirmed, it would be a major breakthrough that should not take years to integrate into mainstream medicine.
Cancer genus (left), dementia genus (right)
Vertical axis: proteins
Horizontal axis: Groups of diseases and specific kinds of disease
in the group, e.g., genus = circulatory system disorders,
species = hypertension, heart failure, etc.
The two figures above from the research article suggests a couple of interesting things. First, this kind of complex AI-driven analysis can be used to predict other kinds of diseases, cancers in the first figure, all kinds of diseases in the second. The point is that AI can comb through gigantic piles of biological data from tens of thousands of people and pick out meaningful bits of information among all the noise of normal, messy biology. I'm not sure humans alone can do that, but if they can, it is much slower going than what AI can do.
Second, age, sex, education and genetics constituted the demographic data this research was based on. Future research can include additional factors for each individual patient, e.g, income, race, etc., to see what other individual factors might increase the predictive accuracy of the blood test. The database this research was based on included a lot more demographic factors than the four listed, but AI just picked those four out of the data obtained from patients when they initially donated blood. This research should point to other data that could be obtained from healthy donors for future rounds of analysis to look for even more accurate predictive tests for all kinds of diseases.
The Communications paper comments reflects those possibilities:
Developing a single-domain assay to identify individuals at high risk of future events is a priority for multi-disease and mortality prevention. By training a neural network, we developed a disease/mortality-specific proteomic risk score (ProRS) based on 1461 Olink plasma proteins measured in 52,006 UK Biobank participants. This integrative score markedly stratified the risk for 45 common conditions, including infectious, hematological, endocrine, psychiatric, neurological, sensory, circulatory, respiratory, digestive, cutaneous, musculoskeletal, and genitourinary diseases, cancers, and mortality. The discriminations witnessed high accuracies achieved by ProRS for 10 endpoints (e.g., cancer, dementia, and death), with C-indexes exceeding 0.80 [exceeding 80% accuracy]. .... Our models were internally validated in the UK Biobank; thus, further independent external validations are necessary to confirm our findings before application in clinical settings.
Molecular medicine is getting to be very interesting.
______________________________________________________________________
______________________________________________________________________
Mitigating climate change necessitates global cooperation, yet global data on individuals’ willingness to act remain scarce. In this study, we conducted a representative survey across 125 countries, interviewing nearly 130,000 individuals. Our findings reveal widespread support for climate action. Notably, 69% of the global population expresses a willingness to contribute 1% of their personal income, 86% endorse pro-climate social norms and 89% demand intensified political action. Countries facing heightened vulnerability to climate change show a particularly high willingness to contribute. Despite these encouraging statistics, we document that the world is in a state of pluralistic ignorance, wherein individuals around the globe systematically underestimate the willingness of their fellow citizens to act. This perception gap, combined with individuals showing conditionally cooperative behavior, poses challenges to further climate action. Therefore, raising awareness about the broad global support for climate action becomes critically important in promoting a unified response to climate change.
The world’s climate is a global common good and protecting it requires the cooperative effort of individuals across the globe. Consequently, the ‘human factor’ is critical and renders the behavioral science perspective on climate change indispensable for effective climate action. Despite its importance, limited knowledge exists regarding the willingness of the global population to cooperate and act against climate change. To fill this gap, we designed and conducted a globally representative survey in 125 countries, with the aim of examining the potential for successful global climate action. The central question we seek to answer is to what extent are individuals around the globe willing to contribute to the common good, and how do people perceive other people’s willingness to contribute (WTC)?
I interpret this to support my belief that there exists (1) world wide dark free speech, coupled with successful political corruption schemes by interests and entities (human and corporate) that profit from pollution and more climate change, and (2) those have been majors factor that have poisoned the collective human mind about climate change. Those are the single most important factors in contributing to our collective ignorance and paralysis.
The data is summarized below.
We have to (1) learn and trust each other or face very bad climate consequences, and (2) understand who the deadly enemies are here, e.g., the staunchly pro-pollution, radical authoritarian Republican Party, oil, gas and coal Cos. like Exxon-Mobil, the plastics and chemical industries, pro-pollution lobbyists, etc., and call them out for their moral and actual crimes (including legalization of formerly illegal polluting activity).