来自用户 颜林林 的文献。
当前共找到 127 篇文献分享,本页显示第 121 - 127 篇。
121.
颜林林 (2022-05-31 07:28):
#paper doi:10.1038/s41586-021-03583-3 Nature 2021, Swarm Learning for decentralized and confidential clinical machine learning. 精准医学的发展得益于数据的快速积累,然而数据共享却始终是数据充分使用的重大障碍。本文提出一种群体学习方法,它结合了边缘计算、区块链等技术,使数据拥有者可以在不违反隐私法规的情况下,让数据可以在全球范围被集成使用,从而解决药物靶标发现、诊断标志物发现等精准医学研究目标所亟需的大规模数据整合需求。为验证该方法的可行性,本文选取了四种疾病,新冠、结核、白血病和肺病,包括血液转录组和胸部X光片数据,且这些数据存在普遍的异质性和对照分布不均匀等问题,对这些数据进行此群体学习的分析。通过将数据分散到不同网络节点,并让这些节点动态加入计算,最终实现对这些疾病的识别或亚型鉴定,并与传统机器学习方法结果进行对比。本文最近在Nature Reviews Immunology的一篇comment上被再次提及,并介绍了其白血病临床诊断已被欧盟成功标准化并随后商业化,进一步验证了该方法的实际价值。同时,由于它以“共享见解而非共享数据(sharing insights, not data)”的方式进行协作,对于当下诸如医学免疫学等复杂研究,也将起到更大作用。
IF:50.500Q1 Nature, 2021-06. DOI: 10.1038/s41586-021-03583-3 PMID: 34040261 PMCID:PMC8189907
Abstract:
Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine. Patients with leukaemia can be identified using machine learning on the basis … >>>
Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine. <<<
翻译
122.
颜林林 (2022-05-29 23:45):
#paper doi:10.1016/j.ccell.2022.05.005 Cancer Cell, 2022, Redefining breast cancer subtypes to guide treatment prioritization and maximize response: Predictive biomarkers across 10 cancer therapies. 这篇论文是关于一项正在进行的、开放标签、适应性、随机II期、多中心的临床试验(NCT01042379)I-SPY2。该临床试验入组高危II期和III期乳腺癌患者,并用70基因MammaPrint测试来排除其中无法从化疗获益的患者,将她们随机分配到10个不同用药组,并在手术前的新辅助治疗期间的不同时间点,进行MRI检查、穿刺取样和/或外周血采集,新辅助治疗前的穿刺样本,同时开展了基因表达芯片、蛋白磷酸化和免疫组化/原位杂交的检测。通过这些检测数据和患者用药响应结果,本研究对入组的987例患者做了重新分类,定义出五种亚型:HER2-/Immune-/DRD-、HER2-/Immune+、HER2-/Immune-/DRD+、HER2+/BP-HER2_or_Basal及HER2+/BP-Luminal。这个重定义过程,除了纳入传统定义所采用的HR/HER2状态外,也包含了诸如增值、DRD、免疫等其他表型特征,在确保分型区分选取最佳治疗方案,即获得最高的pCR(病理完全缓解)的概率,同时也兼顾检测平台稳健性和临床实施简单。虽然目前每个单臂上的病例数还并不算特别多,但相信随着该临床试验的持续开展和更多病例数据的积累,这项研究将优化出相比当前指南建议更好的治疗决策路径。而相应的研究方法和数据分析套路,也预期可以套用到其他癌种上,并在各类高通量多组学检测方法快速发展的今天,持续产出更多精准医疗实践应用。
IF:48.800Q1 Cancer cell, 2022-06-13. DOI: 10.1016/j.ccell.2022.05.005 PMID: 35623341 PMCID:PMC9426306
Abstract:
Using pre-treatment gene expression, protein/phosphoprotein, and clinical data from the I-SPY2 neoadjuvant platform trial (NCT01042379), we create alternative breast cancer subtypes incorporating tumor biology beyond clinical hormone receptor (HR) and … >>>
Using pre-treatment gene expression, protein/phosphoprotein, and clinical data from the I-SPY2 neoadjuvant platform trial (NCT01042379), we create alternative breast cancer subtypes incorporating tumor biology beyond clinical hormone receptor (HR) and human epidermal growth factor receptor-2 (HER2) status to better predict drug responses. We assess the predictive performance of mechanism-of-action biomarkers from ∼990 patients treated with 10 regimens targeting diverse biology. We explore >11 subtyping schemas and identify treatment-subtype pairs maximizing the pathologic complete response (pCR) rate over the population. The best performing schemas incorporate Immune, DNA repair, and HER2/Luminal phenotypes. Subsequent treatment allocation increases the overall pCR rate to 63% from 51% using HR/HER2-based treatment selection. pCR gains from reclassification and improved patient selection are highest in HR subsets (>15%). As new treatments are introduced, the subtyping schema determines the minimum response needed to show efficacy. This data platform provides an unprecedented resource and supports the usage of response-based subtypes to guide future treatment prioritization. <<<
翻译
123.
颜林林 (2022-04-30 18:41):
#paper doi:10.1016/j.ccell.2022.04.002 Cancer Cell, 2022, The translational challenges of precision oncology. 这是一篇新近发表在Cancer Cell上的关于精准肿瘤学(precision oncology)的综述。所谓精准肿瘤学,是指基于肿瘤分子特征进行肿瘤诊治决策。这篇综述回顾了与肿瘤分子特征相关的研究历史和当前研究进展,从肿瘤发生、肿瘤预防、早期检测、新辅助治疗、微小病变残留监测、药物耐受、肿瘤演化过程、肿瘤转移等诊治不同阶段环节,讨论了相应重要分子特征的发现及应用。本文对于目前在肿瘤基因检测行业中涉及到的各类应用,包括涉及的临床队列研究和相关资源,都有提及,整体上内容全面、逻辑脉络清晰。比较适合初学者,快速了解这个方向的产业应用和临床应用,并强烈建议可追溯其参考文献,对各个具体应用场景,进行深入探索和学习。
IF:48.800Q1 Cancer cell, 2022-05-09. DOI: 10.1016/j.ccell.2022.04.002 PMID: 35487215
Abstract:
The translational challenges in the field of precision oncology are in part related to the biological complexity and diversity of this disease. Technological advances in genomics have facilitated large sequencing … >>>
The translational challenges in the field of precision oncology are in part related to the biological complexity and diversity of this disease. Technological advances in genomics have facilitated large sequencing efforts and discoveries that have further supported this notion. In this review, we reflect on the impact of these discoveries on our understanding of several concepts: cancer initiation, cancer prevention, early detection, adjuvant therapy and minimal residual disease monitoring, cancer drug resistance, and cancer evolution in metastasis. We discuss key areas of focus for improving cancer outcomes, from biological insights to clinical application, and suggest where the development of these technologies will lead us. Finally, we discuss practical challenges to the wider adoption of molecular profiling in the clinic and the need for robust translational infrastructure. <<<
翻译
124.
颜林林 (2022-03-20 16:16):
#paper doi:10.1101/2022.03.14.22272390 medRxiv, 2022, AI-Augmented Clinical Decision Support in a Patient-Centric Precision Oncology Registry. 这篇文章介绍了一项由人工智能技术辅助开展肿瘤精准诊疗的工作。文章作者来自xCures公司(成立于2018年),通讯作者Jeff Shrager是该公司的创始人,同时也是斯坦福的客座教授。实现类似目的的医学专家系统,至少可以追溯至上世纪七十年代,到如今即使看似AI技术早已渗透至医疗诸多领域,但却依然缺乏卓有成效的医疗决策通用解决方案,随着IBM Watson折戟,更让人们意识到这个伟大梦想所遭遇的种种现实困难。本文介绍了xCures公司开展的一项临床试验(NCT03793088),该临床试验建立了一个在线平台XCELSIOR,用于登记癌症患者信息,使用NLP等技术对患者病历数据进行格式化和标准化,号称“以患者为中心”,应该是形成类似于患者大健康病历记录,再结合各类公共数据库资源及其他信息,形成针对患者的可选诊疗方案的推荐及排序,并通过由分子药理学家和肿瘤专家组成的虚拟肿瘤委员会 (VTB) 团队进行人工审查,将结果提供给医生和患者,用于指导后续治疗方案决策。该临床试验面向难治性或晚期癌症患者,预计入组1万人(起始于2019年,预计2024年完成),目前已入组2千多人,并以每周约15例患者的速度在持续。其目标正如该公司官网上宣称的“uses artificial intelligence (A.I.) and predictive modeling to identify and rank the most promising treatment options for people with cancer who have exhausted the standard of care”。我个人相信,类似工作在全球各地肯定并非屈指可数,这些工作在未来必然会体现出难以估量的价值,但价值究竟如何体现,目前尚不明朗,即使这篇文章也未呈现出其独特优势所在,临床试验的评判终点也语焉不详,更多信息还有待继续观察。不过从本文及其补充材料的详细介绍看,该工作的工程意义更胜于科学意义,而这篇preprint更多可能是宣传价值。该工作是否会给其公司形成足够盈利也很难说,但能够以正式临床试验的形式开展,看似认真地建立体系并执行,其中细节应该也还是值得关注和学习的。
125.
颜林林 (2022-03-06 20:48):
#paper doi:10.1101/2021.07.19.452956, bioRxiv, 2022, The Tabula Sapiens: a multiple organ single cell transcriptomic atlas of humans. 这是一篇preprint,介绍了对于单细胞转录组测序而言非常重磅的一项资源。它纳入了15位捐赠者(一般由于中风、外伤或缺氧等导致死亡,参见:https://tabula-sapiens-portal.ds.czbiohub.org/whereisthedata)所提供的24个不同组织器官,分离得到将近50万个单细胞,分别进行了10x和/或SmartSeq2的单细胞转录组测序技术,分析得到400多种细胞类型的组织特异性表达数据,提供了组织间T细胞克隆分布、B细胞组织特异性突变率、细胞周期状态及不同细胞在组织器官之间的分布、个体不同组织间细胞类型特异性RNA剪接形式等重要参考基准图谱信息。同时,通过对样本进行病理切片和H&E染色等分析,也将转录组数据与宏观临床相关信息,如不同组织类型的空间异质性、细胞相对丰度估计等都做了关联和讨论。这个项目由 Tabula Sapiens Consortium 执行,其数据(包括原始测序数据和分析结果)存放在AWS、FigShare、CellXGene等平台,供全世界开放使用(但不允许在未征得该委员会及合作方同意前发表图谱或组织规模的数据分析文章),相关信息可在项目网站(https://tabula-sapiens-portal.ds.czbiohub.org/)上找到,该网站还提供了一套流程,帮助用户使用其结果来注释和解读自己的数据。有两点很值得一提:一、该委员会及项目主要由 Chan Zuckerberg Initiative 基金会支持,该基金会由 Facebook创始人马克·扎克伯格及其妻子普莉希拉·陈(生物学专业)共同创办,bioRxiv和medRxiv也是由该基金会支持建立和维持运营;二、这篇文章的通讯作者Stephen R Quake,是生物技术领域的超级大牛,他也应该是在很早期将自己基因组贡献出来验证相关高通量测序技术的名人之一,可参见2009年NBT文章(doi:10.1038/nbt.1561),该文章的受试者P0(猜测很可能就是Quake本人),基于已成为历史的Helicos Biosciences公司的单分子高通量测序技术(应该属于三代测序体系;要知道,二代测序的兴起,也仅仅开始于2008年左右),测定了该技术的最早人全基因组数据。Quake的贡献及事迹这里不做展开,有兴趣者可自行搜索。
126.
颜林林 (2022-02-28 21:05):
#paper doi:10.1101/mcs.a006198 Cold Spring Harbor Molecular Case Studies, 2022, Towards transcriptomics as a primary tool for rare disease investigation. 来自斯坦福的一篇小综述,介绍转录组测序(RNA-seq)近年在罕见病研究中的应用及进展。选择并推荐这篇paper,是因为一年多以前,有一次我跟一位医生沟通课题思路时,曾经探讨过这种可能性,由于RNA相比DNA更接近于(生物分子行使)功能,所以优先选择RNA-seq,比DNA-seq可能更具探索价值。通常,疾病的遗传学研究和分子诊断,都会在基因组(DNA)层面进行。然而在罕见病中,通过基因组测序得到的诊断率并不高,且很多基因组突变也面临注释解读的困难,因而近年开始尝试辅以RNA-seq。RNA-seq虽然不能准确代表遗传背景,但它在基因表达量变化、可变剪接、外显子跳跃,内含子保留等诸多方面,却能提供更多与疾病或表型相关的信息。在最新的大型罕见病队列中,RNA-seq也的确提升了诊断率。随着相关研究的开展,与罕见病RNA-seq相关的计算方法也有所改进,比如针对罕见病患者与对照人群比例不平衡的统计模型,再比如整合DNA与RNA测序数据的等位基因间差异表达。虽然转录组的确会受限于样本类型和采样时间,很容易从一开始就遗漏特定组学变异事件,但最新的一些研究,通过将样本细胞转分化为其他特定细胞,从而发现与疾病相关的表达基因及其突变事件,也为RNA-seq在疾病研究中的应用提供了拓展思路。目前还没有关于RNA-seq诊断率的随机临床试验,因此这些研究都还只是探索性或回顾性的,不过从所取得的进展看,RNA-seq这项经典技术的更大价值发挥,未来还是很值得期待的。
Abstract:
In the past 5 years transcriptome or RNA-sequencing (RNA-seq) has steadily emerged as a complementary assay for rare disease diagnosis and discovery. In this perspective, we summarize several recent developments … >>>
In the past 5 years transcriptome or RNA-sequencing (RNA-seq) has steadily emerged as a complementary assay for rare disease diagnosis and discovery. In this perspective, we summarize several recent developments and challenges in the use of RNA-seq for rare disease investigation. Using an accessible patient sample, such as blood, skin, or muscle, RNA-seq enables the assay of expressed RNA transcripts. Analysis of RNA-seq allows the identification of aberrant or outlier gene expression and alternative splicing as functional evidence to support rare disease study and diagnosis. Further, many types of variant effects can be profiled beyond coding variants, as the consequences of noncoding variants that impact gene expression and splicing can be directly observed. This is particularly apparent for structural variants that disproportionately underlie outlier gene expression and for splicing variants in which RNA-seq can both measure aberrant canonical splicing and detect deep intronic effects. However, a major potential limitation of RNA-seq in rare disease investigation is the developmental and cell type specificity of gene expression as a pathogenic variant's effect may be limited to a specific spatiotemporal context and access to a patient's tissue sample from the relevant tissue and timing of disease expression may not be possible. We speculate that as advances in computational methods and emerging experimental techniques overcome both developmental and cell type specificity, there will be broadening use of RNA sequencing and multiomics in rare disease diagnosis and delivery of precision health. <<<
翻译
127.
颜林林 (2022-01-16 13:16):
#paper doi:10.3322/caac.21708 CA: A Cancer Journal for Clinicians, 2022, Cancer statistics, 2022。这是最新发表的美国癌症统计数据,汇编了截至2018年的发病率数据及截至2019年的死亡率数据,并对其趋势进行预测和分析。主要结论是:乳腺癌和前列腺癌的进展停滞不前,但肺癌的进展却有所加强。CA杂志上每隔几年就会有关于世界范围或国家范围的癌症流调结果文章发表,算是重要的专业数据源及其解读,值得关注和阅读。值得注意的一句话:疫情导致医疗机构关闭或因恐惧暴露而减少护理,导致诊断和治疗延误,可能导致癌症发病率短期下降,随后晚期疾病上升,并最终增加死亡,相关数据收集需要滞后数年时间。
癌症统计,2022
Abstract:
Each year, the American Cancer Society estimates the numbers of new cancer cases and deaths in the United States and compiles the most recent data on population-based cancer occurrence and … >>>
Each year, the American Cancer Society estimates the numbers of new cancer cases and deaths in the United States and compiles the most recent data on population-based cancer occurrence and outcomes. Incidence data (through 2018) were collected by the Surveillance, Epidemiology, and End Results program; the National Program of Cancer Registries; and the North American Association of Central Cancer Registries. Mortality data (through 2019) were collected by the National Center for Health Statistics. In 2022, 1,918,030 new cancer cases and 609,360 cancer deaths are projected to occur in the United States, including approximately 350 deaths per day from lung cancer, the leading cause of cancer death. Incidence during 2014 through 2018 continued a slow increase for female breast cancer (by 0.5% annually) and remained stable for prostate cancer, despite a 4% to 6% annual increase for advanced disease since 2011. Consequently, the proportion of prostate cancer diagnosed at a distant stage increased from 3.9% to 8.2% over the past decade. In contrast, lung cancer incidence continued to decline steeply for advanced disease while rates for localized-stage increased suddenly by 4.5% annually, contributing to gains both in the proportion of localized-stage diagnoses (from 17% in 2004 to 28% in 2018) and 3-year relative survival (from 21% to 31%). Mortality patterns reflect incidence trends, with declines accelerating for lung cancer, slowing for breast cancer, and stabilizing for prostate cancer. In summary, progress has stagnated for breast and prostate cancers but strengthened for lung cancer, coinciding with changes in medical practice related to cancer screening and/or treatment. More targeted cancer control interventions and investment in improved early detection and treatment would facilitate reductions in cancer mortality. <<<
翻译
每年,美国癌症协会都会估计美国新发癌症病例和死亡人数,并汇编有关基于人群的癌症发病率和结果的最新数据。发病率数据(至2018年)由监测、流行病学和最终结果计划收集;国家癌症登记计划;以及北美中央癌症登记协会(North American Association of Central Cancer Registries)。死亡率数据(截至 2019 年)由国家卫生统计中心收集。2022 年,预计美国将发生 1,918,030 例新发癌症病例和 609,360 例癌症死亡,其中每天约有 350 人死于肺癌,肺癌是癌症死亡的主要原因。2014 年至 2018 年期间,女性乳腺癌的发病率继续缓慢增加(每年增长 0.5%),前列腺癌的发病率保持稳定,尽管自 2011 年以来晚期乳腺癌的发病率每年增加 4% 至 6%。因此,在过去十年中,诊断为晚期前列腺癌的比例从3.9%增加到8.2%。相比之下,晚期肺癌的发病率继续急剧下降,而局限性肺癌的发病率每年突然上升4.5%,有助于提高局限性诊断的比例(从2004年的17%上升到2018年的28%)和3年相对生存率(从21%上升到31%)。死亡率模式反映了发病趋势,肺癌的下降速度加快,乳腺癌的下降速度减慢,前列腺癌的下降趋势趋于稳定。总之,乳腺癌和前列腺癌的进展停滞不前,但肺癌的进展有所加强,这与与癌症筛查和/或治疗相关的医疗实践的变化相吻合。更有针对性的癌症控制干预措施和对改进早期发现和治疗的投资将有助于降低癌症死亡率。
回到顶部