Believe in science

Health

Antimicrobial resistance has become one of the leading causes of death worldwide.

Antimicrobial drugs (Antimicrobial), including antibiotics, antiviral drugs, antifungal drugs and antiparasitic drugs, are used to prevent and treat human, animal and plant infections. However, these microbes can develop resistance and resistance to the drugs that treat them, that is, antimicrobial resistance (Antimicrobial resistance,AMR). Recently, a study published in the Lancet said that AMR has become one of the leading causes of global death and may be more serious than the threat posed by HIV or malaria.

The researchers counted the number of AMR-related deaths and disability-adjusted life years (DALYs) in 2019, involving 23 pathogens and 88 pathogen-drug combinations. After analyzing 471 million records, it was found that 4.95 million deaths in 2019 were related to drug-resistant bacterial infections, of which 1.27 million were directly caused by AMR. Among the pathogens, Escherichia coli caused the largest number of deaths, while in the pathogen-drug combination, methicillin-resistant Staphylococcus aureus caused the highest number of deaths-more than 100000. The study is considered to be the most comprehensive assessment of AMR to date.

Information technology.

Interactive data may be used to identify anonymous individuals for a long time.

Now, mobile programs such as communication applications collect user interaction data, which can be used to develop better products or carry out related research. Under current data protection regulations, the data can be shared and sold without user consent, but only if the data is anonymous. But a study published yesterday in the journal Nature Communications found that these interactive data can be used to identify individuals even if they are anonymous.

The researchers collected anonymous information about more than 40,000 mobile phone users, mainly about their interactions with others on communication apps and other platforms. They also developed a model based on deep learning and used it to analyze the collected data to find individual interaction patterns. The results show that the model can identify 15% of the people based on the individual's direct (level 1) interactive network, and 52% of the people based on the individual's level 2 interactive network (the individual interaction with one person separated from the target individual). In addition, 20 weeks later, the researchers were able to identify 24% of people using an individual's level 2 interactive network. This shows that the current anonymized interactive data has long-term identifiability, so the access to such data should be limited.

Astronomy

Discovery of intermediate mass black holes in neighboring galaxies

At present, astronomers have successfully detected supermassive black holes in the center of galaxies and low-mass black holes produced by star explosions, but medium-mass black holes are difficult to find. A study published in the Astrophysical Journal on January 11 found a medium-mass black hole about 100000 times the mass of the sun in the B023-G078 cluster in the Andromeda galaxy.

Using new observations from the Gemini Observatory and images from the Hubble Space Telescope, the researchers plotted the B023-G078 profile, calculated the mass distribution within the cluster, and speculated the speed of stars at various positions in the cluster. By comparing the modeling analysis with the actual observation results, it is found that the motion of the star in the center of the model without black hole is slower than that of observation, while in the model of medium mass black hole, the star motion velocity is consistent with the observation. In addition, celestial bodies in globular clusters are formed almost at the same time, and the light contours of their inner and outer regions are usually similar. On the other hand, the light profile of the center of the cluster is close to a circle, while the outer region tends to be flattened, and the chemical composition of stars in different regions is also different, and the central star contains more heavy elements. This suggests that B023-G078 may be the exfoliated nucleus of a small galaxy falling into the Andromeda galaxy. The researchers plan to observe more exfoliated nuclei of small galaxies and look for traces of medium-mass black holes to deepen their understanding of galaxy formation and merger.

Medical science

Using the eye as a "window" to monitor cardiovascular disease

Cardiovascular disease is one of the main causes of human premature death worldwide, but at present, diagnostic techniques such as cardiac ultrasound and nuclear magnetic resonance are expensive and can only be carried out in hospitals. On the other hand, changes in tiny blood vessels in the retina can be used as an indication of more general vascular diseases, including heart disease. A new study published in Natural Machine Intelligence has developed an artificial intelligence (AI) system based on deep learning algorithms that can identify people at high risk of heart disease through retinal scans.

The researchers trained the AI system to analyze retinal and cardiac scans of more than 5000 people to identify the link between retinopathy and heart changes in patients. After training, the system can estimate left ventricular size and end-diastolic volume only from retinal scans, and ventricular hypertrophy is associated with an increased risk of heart disease. Then combined with basic demographic information such as age and gender, the AI system can predict the risk of myocardial infarction in the coming year, with an accuracy of between 70% and 80%. Researchers say retinal scans are relatively cheap and routine eye examinations, and such an AI system is expected to improve early heart disease screening and help high-risk people see a doctor in a timely manner.

Health

The two sides of human longevity genes

The scientists analyzed genetic data from 11,262 long-lived people and found that people were more likely to live longer if the activity of specific genes associated with RNA polymerase (Pol I, Pol III) and ribosomal protein expression was lower. Past studies have also shown that drugs inhibiting the activity of Pol III may help prolong life. These genes are necessary for the synthesis of key proteins in cells, so they are essential in the early stages of development, but can cause disease later in life. This shows that these genes seem to have antagonistic pleiotropy, that is, one gene affects many traits at the same time, and although some of these traits are disadvantageous to the individual in the later stage of life, others are beneficial to the early stage of individual development. as a result, the gene can be preserved by natural selection.

Physics

B meson decay conforms to the standard model prediction

B meson is an unstable neutral or charged particle composed of an antibottom quark and another quark. Several decay modes of B meson involve the exchange of multiple virtual particles, so it is very rare. If accurate measurements show that the rates of these decays are not consistent with those predicted by the standard model, then there may be new particles or physical processes outside the standard model. Yesterday, the large Hadron Collider bottom Quark experiment (LHCb) cooperative published its measurements of three rare decays in the physical Review KuaiBao, showing that the decay rate of B mesons is in line with the prediction of the standard model.

The team used LHCb detectors to measure the decay of bottom quarks produced during proton-proton collisions. They carried out experiments at 7, 8 and 13 TeV centroid collision energies, respectively. It is found that the strange B meson (composed of anti-bottom quark and strange quark) will decay into a pair of muons with a probability of 3 × 10-9, which is consistent with the previous measurements. No other two kinds of decay have b