As aneuroscientist surveying the landscape of generative AI—artificial intelligence capable of generating text,images,or other media—Konrad Kording cites two potential directions forward:One is the“weird future”of political use and manipulation,and the other is the“power tool direction,”where people use ChatGPT to get information as they would use adrill to build furniture.“I’m not sure which of those two directions we’re going but Ithink alot of the AI people are working to move us into the power tool direction,”says Kording,a Penn Integrates Knowledge(PIK)University professor with appointments in the Perelman School of Medicine and School of Engineering and Applied Science.Reflecting on how generative AI is shifting the paradigm of science as adiscipline,Kording said he thinks“it will push science as awhole into amuch more collaborative direction,”though he has concerns about ChatGPT’s blind spots.Kording joined three University of Pennsylvania researchers from the chemistry,political science,and psychology departments sharing their perspectives in the recent panel“ChatGPT turns one:How is generative AI reshaping science?”PIK Professor RenéVidal opened the event,which was hosted by the School of Arts&Sciences’Data Driven Discovery Initiative(DDDI),and Bhuvnesh Jain,physics and astronomy professor and co-faculty director of DDDI,moderated the discussion.“Generative AI is moving so rapidly that even if it’s asnapshot,it will be very interesting for all of us to get that snapshot from these wonderful experts,”Jain said.OpenAI launched ChatGPT,a large language model(LLM)-based chatbot,on Nov.30,2022,and it rapidly ascended to ubiquity in news reports,faculty discussions,and research papers.Colin Twomey,interim executive director of DDDI,told Penn Today that it’s an open question as to how it will change the landscape of scientific research,and the`idea of the event was to solicit colleagues’opinions on interesting directions in their fields.In honor of what he called“ChatGPT’s birthday party,”assistant professor of chemistry Andrew Zahrt asked the chatbot about its use in his field and got the response,“Generative AI in chemistry serves as acreative digital assistant,helping scientists design new molecules and materials by predicting their properties and suggesting innovative combinations,ultimately accelerating the drug discovery and materials development processes.”Zahrt called that“a really good description.”He said when it comes to generative AI,the chemistry field is still in the proof of principle stage,where alot of research is asking whether scientists can use generative models to propose reasonable chemical structures. 查看详细>>

Autoimmune disorders are among the most prevalent chronic diseases across the globe.Emerging treatments for autoimmune disorders focus on“adoptive cell therapies,”or those using cells from apatient’s own body to achieve immunosuppression.These therapeutic cells are recognized by the patient’s body as“self,”therefore limiting side effects,and are specifically engineered to localize the intended therapeutic effect.In treating autoimmune diseases,current adoptive cell therapies have largely centered around the regulatory Tcell(Treg).Although Tregs offer great potential,using them for therapeutic purposes remains amajor challenge.In particular,current delivery methods result in inefficient engineering of Tcells.Tregs only compose approximately 5-10%of circulating peripheral blood mononuclear cells.Furthermore,Tregs lack more specific surface markers that differentiate them from other Tcell populations.These hurdles make it difficult to harvest,purify,and grow Tregs to therapeutically relevant numbers.Now,a research team led by Michael Mitchell,associate professor in Bioengineering in the School of Engineering and Applied Science,has developed alipid nanoparticle(LNP)platform to deliver Foxp3 messenger RNA(mRNA)to Tcells for applications in autoimmunity.Their findings are published in the journal Nano Letters.“The major challenges associated with ex vivo(outside the body)cell engineering are efficiency,toxicity,and scale-up:our mRNA lipid nanoparticles(mRNA LNPs)allow us to overcome all of these issues,”says Mitchell.“Our work’s novelty comes from three major components:first,the use of mRNA,which allows for the generation of transient immunosuppressive cells;second,the use of LNPs,which allow for effective delivery of mRNA and efficient cell engineering;and last,the ex vivo engineering of primary human Tcells for autoimmune diseases,offering the most direct pipeline for clinical translation of this therapy from bench to bedside.”“To our knowledge,this is one of the first mRNA LNP platforms that has been used to engineer Tcells for autoimmune therapies,”Mitchell says.“Broadly,this platform can be used to engineer adoptive cell therapies for specific autoimmune diseases and can potentially be used to create therapeutic avenues for allergies,organ transplantation and beyond.” 查看详细>>

As artificial intelligence becomes more integrated into our daily lives,it’s essential that these systems are able to accurately make decisions in the real world,and react appropriately in complex,ever-changing environments.That’s the goal of research by Kaustubh Sridhar,a doctoral student in the Penn Research in Embedded Computing and Integrated Systems Engineering(PRECISE)Center in the School of Engineering and Applied Science.Sridhar’s research on AI decision-making aims to make current models of the world more accurate,to create AI systems that can better learn from expert data,and to create systems that can adapt to new situations quickly.“I would like to continue doing AI and decision-making research,possibly in industry,after Igraduate so that Ican help improve autonomous agents in the real world,”says Sridhar.“And of course,my hope is that such agents will help people in awide range of tasks ranging from household chores to autonomous driving.”“In undergrad,I spent my efforts inventing control algorithms for quadrotors,robots with four rotors,which Ireally liked,”says Sridhar.“I later expanded my boundaries to discover that generally intelligent decision making is the most challenging problem for obtaining robots that can help humans in the real world.”Throughout his doctoral studies,Sridhar has been working to build general algorithms that are both intelligent and safe for any robot and any decision-making agent.He is atwo-time recipient of Outstanding Reviewer awards in top machine learning conferences.His work was nominated for the Best Paper award at the International Conference on Cyber-Physical Systems(ICCPS)in 2023.His most recent research has focused on semi-parametric methods for machine learning and decision-making applications that can be used for robot learning and learning-enabled digital or physical systems,such as autonomous vehicles,electric grids,and cloud computing.“The key takeaways include that simple ways to combine the benefits of non-parametric components with neural networks leads to better generalization in learning both dynamics models and policies,”he says.“Some conclusions that surprised me were that semiparametric methods can provide rigorous guarantees for real-world performance in diverse domains that otherwise cannot be found.” 查看详细>>

New research finds that driving skills measured at the time of licensure on avirtual driving assessment(VDA),which exposes drivers to common serious crash scenarios,helps predict crash risk in newly licensed teen drivers.Published in the journal Pediatrics,and conducted by the Center for Injury Research and Prevention(CIRP)at Children’s Hospital of Philadelphia(CHOP)with colleagues at Penn’s Annenberg Public Policy Center and the University of Michigan,the study brings the research community one step closer to identifying which skill deficits put young new drivers at higher risk for crashes.With this cutting-edge information,more personalized interventions can be developed to improve the driving skills that prevent crashes.Image:iStock/chombosan While drivers between the ages of 15 and 20 only make up about 5%of all drivers on the road,they are involved in approximately 12%of all vehicle crashes and 8.5%of fatal crashes.The time of greatest crash risk is in the months right after these young drivers receive their license,largely due to deficits in driving skills.However,many of these newly licensed drivers do avoid crashes.The challenge for policymakers,clinicians,and families has been identifying which drivers are at increased risk of crashing during the learning phase before they drive on their own.Early identification of at-risk drivers offers the opportunity to intervene with training and other resources known to help prevent crashes,making the roads safer for everyone.The researchers examined the ability of the VDA,delivered at the time of the licensing road test,to predict crash risk in the first year after obtaining licensure in the state of Ohio.Using aunique study design,the results of the VDA were linked to police-reported crash records for the first year after obtaining alicense.“Our previous research showed that performance on the VDA predicted actual on-road driving performance,as measured by failure on the licensing road test.This new study went further to determine whether VDA performance could identify unsafe driving performance predictive of future crash risk,”says lead study author Elizabeth Walshe,a cognitive neuroscientist and clinical researcher who directs the Neuroscience of Driving team at CIRP.“We found that drivers categorized by their performance as having major issues with dangerous behavior were at higher risk of crashing than average new drivers,”says Walshe,a former postdoctoral fellow at the Annenberg Public Policy Center(APPC).“These findings are incredibly important because they provide us with quantitative evidence that we can approach young driver safety in anew way—by predicting crash risk and aiming resources to those who need them most,”says Flaura Winston,co-scientific director of CIRP at CHOP,co-author of the study,and an APPC distinguished research fellow.“By providing this information before licensure,we can direct resources to those most at risk,and potentially prevent crashes from occurring when these teens first drive on their own.” 查看详细>>

Continuous wrist temperature monitoring can uncover insights into the potential for future disease risk for ailments like Type 2diabetes,hypertension,liver disease,kidney failure,and more.These new findings from Perelman School of Medicine researchers,published in Nature Communications,shows that accurate,continued digital monitoring of skin temperature can give deeper medical insights.Previously,disrupted temperature rhythms had only been linked to ahandful of conditions,such as metabolic syndrome and diabetes.Now,this research provides insights from alarge population,and it indicates awider spectrum of conditions are associated with poor temperature rhythms,measured in wrist temperature amplitude(the difference between the minimum and maximum temperature over the course of 24 hours).“These findings indicate the potential to marry emerging technology with health monitoring in apowerful new way,”says Carsten Skarke,an adjunct associate professor of medicine,Robert L.McNeil Jr.fellow in translational medicine and therapeutics,and the study’s senior author.“For example,there are many who have smart watches around their wrists,which already include skin temperature sensors.In the future,this information may be leveraged with their care teams as adigital biomarker,to understand their risk to develop certain diseases and to navigate treatment or preventative care options.”The findings indicate that the daily peaks and valleys observed in one’s wrist temperature curve might matter for health.The flatter this landscape becomes,the higher the risk for chronic diseases. 查看详细>>

Adults with atopic dermatitis(AD)have a34%increased risk of developing new-onset inflammatory bowel disease(IBD)compared with individuals who do not have the skin condition,and children have a44%increased risk,according to anew study from the Perelman School of Medicine.Additionally,as the severity of AD increased,the risk of developing IBD rose.These findings clear up ambiguity from previous research,especially among populations of children and between the different types of inflammatory bowel disease:ulcerative colitis and Crohn’s disease.Insight offered from this study,published in JAMA Dermatology,could lead to new treatments for both IBD and AD.IBD encompasses the diseases ulcerative colitis and Crohn’s disease,which are disorders involving chronic digestive tract inflammation.While IBD is located in the gut and AD affects the skin,both diseases are driven by the immune system and are categorized by severe inflammation.“It is imperative for clinicians to understand atopic dermatitis and the trajectory of our patients with it in order to provide the best standard of care,”says senior author Joel MGelfand,the James J.Leyden,M.D.Endowed Professor in Clinical Investigation in the Department of Dermatology at Penn.“There are new and better treatments for AD today,and there will likely continue to be more.But providers have to understand how those treatments could impact other autoimmune diseases.For patients with AD and another autoimmune disease,some currently available medications can exacerbate symptoms of their other disease or can help treat two immune diseases at the same time.” 查看详细>>

New research is bolstering scientific understanding behind why some people are more prone to allergies than others.Researchers in Penn’s Perelman School of Medicine identified how genetic differences that alter aspecific protein called ETS1 can affect our body’s response to allergies.They found that small changes in ETS1 in an animal model can lead to an increased likelihood for allergic reactions that cause inflammation.The findings were published recently in Immunity.The United States Centers for Disease Control and Prevention reports that allergies rank as the sixth most prevalent cause of chronic illness in the U.S.,resulting in an annual expenditure exceeding$18 billion.While previous research has established astrong genetic basis for allergies and identified specific genetic sequence variations which predispose for these chronic diseases,how our DNA can affect our chances of developing an allergy remains unclear.But understanding this could lead to improved research and potential new treatments.By using modern genomics and imaging techniques,a collaborative team of researchers co-led by Penn’s Golnaz Vahedi,an associate professor of genetics,and Jorge Henao-Mejia,an associate professor of pathology and laboratory medicine,found that the ETS1 protein plays arole in controlling atype of immune cell called CD4+T helper cells,which are important in allergic reactions and help orchestrate the immune response by activating and coordinating other immune cells.DNA interactions within the genomic segment encompassing the ETS1 gene control how much of the ETS1 protein is made.“We discovered that these interactions,work like adimmer switch,”says Vahedi.“When there are changes in the DNA in this area,it can mess up the dimmer switch,causing problems with controlling the ETS1 protein.This can lead to imbalances in our immune cells and cause allergic inflammations." 查看详细>>

Penn Medicine has aremarkable track record developing therapies that receive FDA approvals,including pioneering whole classes such as cellular and gene therapies and mRNA vaccines.Even so,faculty remain humbled by one basic tenet:Advances mean nothing if patients can’t access them.“If our goal is to improve human health,we cannot call ourselves successful unless we bring the fruits of our labor to bear as broadly as possible on people suffering from the conditions that interventions are designed to treat,”says Emma Meagher,senior vice dean for clinical and translational research in the Perelman School of Medicine.Commercialized drug prices are beyond an academic institution’s control,but researchers—including many at Penn—are exploring advances that would make cutting-edge treatments more easily available to everyone who needs them.In the case of CAR T,for example,those working to develop auniversal version of the therapy understand they can drive down costs by transforming atime-intensive,personalized treatment that requires apatient to donate their own Tcells into one where an existing bank of cells could be used to treat many patients.Racial and ethnic health disparities also stoke accessibility concerns.Black Americans are less likely than any other group to receive CAR Tcell therapy—a trend rooted in gaps in income,education,housing,job security,and proximity to high-quality medical centers.These same social determinants of health have also led to their underrepresentation in CAR Tand all kinds of clinical trials;Black patients account for just 5%of clinical trial participants nationwide.“Black patients’historically low participation in trials means that,compared to white individuals,they have less access to some of the most advanced new treatments,”says Carmen Guerra,vice chair of diversity and inclusion in the Department of Medicine and associate director of diversity and outreach at the Abramson Cancer Center(ACC).To better diversify clinical studies in cancer,the ACC recently conducted afive-year community engagement study that reached more than 10,000 individuals through marketing campaigns tailored to minority cultures;wellness forums and events in Black communities;partnerships with Lyft and Ride Health to reduce transportation barriers;and patient education efforts.By the end of the project,the number of Black patients in Penn’s cancer clinical trials had doubled.Even when drugs are relatively affordable and easy to produce,racial disparities persist—disparities Penn Medicine is striving to overcome.Following the approval of mRNA vaccines against COVID-19,for instance,the health system worked with partners in West and Southwest Philadelphia to bring vaccines directly to their predominantly Black communities in spaces like gyms and public schools,and later moved portable“hyperlocal”clinics between locations like fast-food restaurants and retail parking lots in neighborhoods where many residents remained unvaccinated. 查看详细>>

mammalian DNA is folded in 3D structures that create different“neighborhoods”in the genome.These sections of DNA,formally called“topologically associating domains,”remain insulated from each other in order to control how genes get expressed.But when apiece of DNA in one neighborhood is required to control and develop aunique set of genes in another,the neighborhoods must then intermingle.According to astudy led by Golnaz Vahedi at the Perelman School of Medicine,one protein,called TCF-1,allows various parts of these otherwise insulated DNA to mix in way that’s required for the Tcells—a key element of the body’s immune system—to develop.The role this protein plays in Tcell creation could shed new light on immunotherapy approaches.The team published its findings in Nature Immunology.By studying the mechanics of the protein TCF-1 and how it reconfigures the genome,Vahedi,an associate professor of genetics and amember of the Penn Institute for Immunology and Penn Epigenetics Institute,and colleagues,discovered that the TCF-1 protein has aunique ability to enable plasticity in cells across neighborhoods during the development of Tcells.“These domains,or insulated neighborhoods,are like stickers for social distancing,”Vahedi says.“They essentially say,‘Stay away—keep acertain distance apart.’But what this protein does is to remove these stickers and say,‘You can now actually intermingle.’It disrupts the spatial distancing.” 查看详细>>

Chimeric antigen receptor(CAR)T cells can be remarkably effective in treating leukemias and lymphomas,but there are no successful immunotherapies for neuroendocrine tumors(NETs)and gastrointestinal cancers(GICs)yet.Researchers at Penn Medicine have discovered that CAR Tcells directed to atumor antigen,CDH17,a cell surface marker expressed on both NETs and GICs but also found on healthy tissues,eliminated GICs in several preclinical models without toxicity to normal tissues in multiple mouse organs,including the small intestine and colon.The results from this study,the first to target CDH17 in neuroendocrine tumors,suggest anew class of tumor associated antigens accessible to CAR Tcells in tumors but sequestered from CAR Tcells in healthy tissues.In the study,published in Nature Cancer,researchers isolated allama-derived nanobody,a small antibody,which led to the identification of CDH17.Targeting CDH17—which is mainly expressed in the intestinal system—with CDH17CAR Tcells eliminated gastric,pancreatic,and colorectal cancers in animal models.While CDH17 is also expressed in normal intestinal epithelial cells,the CDH17CAR Tcells did not attack the normal cells,likely because the CAR Tcells cannot reach or bind to healthy tissue in the tight junction between normal intestine epithelial cells,creating a“masking”effect in healthy cells from CAR Tattack.“Our work demonstrates that CDH17CAR Tcells can eliminate solid tumors like NETs and GICs,but do not damage healthy,normal tissues that also express CDH17,because CDH17 is sequestered and hidden between the normal cells,”says senior author Xianxin Hua,a professor in the Department of Cancer Biology in the Abramson Family Cancer Research Institute at Penn.“This opens avenues to explore anew class of tumor antigens that are also expressed in normal cells but protected by the CAR Tcell attack,and is hopefully another important step in developing safer immunotherapies for solid tumors.”There are about five million new cases of GICs annually worldwide,underscoring the need for scientific and clinical advancements. 查看详细>>