抗原变异可使得多种致病微生物易于逃避宿主免疫应答。表达在感染红细胞表面的恶性疟原虫红细胞表面蛋白1（PfPMP1）与感染红细胞、内皮细胞、树突状细胞以及胎盘的单个或多个受体作用，在黏附及免疫逃避中起关键的作用。每个单倍体基因组var基因家族编码约60种成员，通过启动转录不同的var基因变异体为抗原变异提供了分子基础。

疟原虫var基因转换速率变化导致抗原变异[英]／Paul H, Robert P, Christodoulou Z, et al//Proc Natl Acad Sci U S A

Machines are increasingly doing “intelligent” things. Face recognition algorithms use a large dataset of photos labeled as having a face or not to estimate a function that predicts the pre...

https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.31.2.87

Machine Learning: An Applied Econometric Approach

The emergence of artificial intelligence (AI) and its progressively wider impact on many sectors requires an assessment of its effect on the achievement of the Sustainable Development Goals. Using ...

/pdf/the-role-of-artificial-intelligence-in-achieving-the-2cr515vsr6.pdf

The role of artificial intelligence in achieving the Sustainable Development Goals

Lead Authors: Marco Bindi (Italy), Sally Brown (UK), Ines Camilloni (Argentina), Arona Diedhiou (Ivory Coast/Senegal), Riyanti Djalante (Japan/Indonesia), Kristie L. Ebi (USA), Francois Engelbrecht (South Africa), Joel Guiot (France), Yasuaki Hijioka (Japan), Shagun Mehrotra (USA/India), Antony Payne (UK), Sonia I. Seneviratne (Switzerland), Adelle Thomas (Bahamas), Rachel Warren (UK), Guangsheng Zhou (China)

Impacts of 1.5°C Global Warming on Natural and Human Systems

Distribution shifts -- where the training distribution differs from the test distribution -- can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity, these real-world distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated collection of 8 benchmark datasets that reflect a diverse range of distribution shifts which naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training results in substantially lower out-of-distribution than in-distribution performance, and that this gap remains even with models trained by existing methods for handling distribution shifts. This underscores the need for new training methods that produce models which are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. Code and leaderboards are available at this https URL.

WILDS: A Benchmark of in-the-Wild Distribution Shifts

Reliable data on economic livelihoods remain scarce in the developing world, hampering efforts to study these outcomes and to design policies that improve them. Here we demonstrate an accurate, inexpensive, and scalable method for estimating consumption expenditure and asset wealth from high-resolution satellite imagery. Using survey and satellite data from five African countries—Nigeria, Tanzania, Uganda, Malawi, and Rwanda—we show how a convolutional neural network can be trained to identify image features that can explain up to 75% of the variation in local-level economic outcomes. Our method, which requires only publicly available data, could transform efforts to track and target poverty in developing countries. It also demonstrates how powerful machine learning techniques can be applied in a setting with limited training data, suggesting broad potential application across many scientific domains.

Combining satellite imagery and machine learning to predict poverty

International climate change agreements typically specify global warming thresholds as policy targets
 1
 , but the relative economic benefits of achieving these temperature targets remain poorly understood2,3. Uncertainties include the spatial pattern of temperature change, how global and regional economic output will respond to these changes in temperature, and the willingness of societies to trade present for future consumption. Here we combine historical evidence
 4
 with national-level climate
 5
 and socioeconomic
 6
 projections to quantify the economic damages associated with the United Nations (UN) targets of 1.5 °C and 2 °C global warming, and those associated with current UN national-level mitigation commitments (which together approach 3 °C warming
 7
 ). We find that by the end of this century, there is a more than 75% chance that limiting warming to 1.5 °C would reduce economic damages relative to 2 °C, and a more than 60% chance that the accumulated global benefits will exceed US$20 trillion under a 3% discount rate (2010 US dollars). We also estimate that 71% of countries—representing 90% of the global population—have a more than 75% chance of experiencing reduced economic damages at 1.5 °C, with poorer countries benefiting most. Our results could understate the benefits of limiting warming to 1.5 °C if unprecedented extreme outcomes, such as large-scale sea level rise
 8
 , occur for warming of 2 °C but not for warming of 1.5 °C. Inclusion of other unquantified sources of uncertainty, such as uncertainty in secular growth rates beyond that contained in existing socioeconomic scenarios, could also result in less precise impact estimates. We find considerably greater reductions in global economic output beyond 2 °C. Relative to a world that did not warm beyond 2000–2010 levels, we project 15%–25% reductions in per capita output by 2100 for the 2.5–3 °C of global warming implied by current national commitments
 7
 , and reductions of more than 30% for 4 °C warming. Our results therefore suggest that achieving the 1.5 °C target is likely to reduce aggregate damages and lessen global inequality, and that failing to meet the 2 °C target is likely to increase economic damages substantially.

W. Matthew Davis

Papers

Combining satellite imagery and machine learning to predict poverty

Large potential reduction in economic damages under UN mitigation targets.