Chen Junfang, Guest Paul C, Schwarz Emanuel
Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany.
Laboratory of Neuroproteomics, Department of Biochemistry and Tissue Biology, Institute of Biology, University of Campinas (UNICAMP), Rua Monteiro Lobato 255 F/01, Cidade Universitária Zeferino Vaz, 13083-862, Campinas, Brazil.
Adv Exp Med Biol. 2017;974:131-138. doi: 10.1007/978-3-319-52479-5_8.
As substantial efforts are being made to identify biological markers of psychiatric illnesses, it is becoming clear that clinically useful accuracy will require larger sets of readouts that potentially span different technological platforms. For discovery of proteomic biomarkers, simultaneous measurement of analytes in small sample quantities has developed into a widely used technology of similar quality as the respective single-plex assays. Multiplex assay systems therefore hold promise for biomarker discovery and development in many complex disease areas including psychiatry. However, analysis of the derived data is subject to substantial challenges that may impede the possibility of obtaining meaningful findings. This chapter discusses potential applications of multiplexed assay technologies during biomarker development and highlights potential challenges for machine learning analysis of derived data.
随着人们为确定精神疾病的生物标志物付出了巨大努力,越来越明显的是,临床上有用的准确性将需要更大量的读数集,这些读数可能跨越不同的技术平台。对于蛋白质组学生物标志物的发现,对少量样品中的分析物进行同步测量已发展成为一种广泛使用的技术,其质量与各自的单重分析方法相当。因此,多重分析系统在包括精神病学在内的许多复杂疾病领域的生物标志物发现和开发方面具有前景。然而,对所得数据的分析面临重大挑战,这可能会阻碍获得有意义结果的可能性。本章讨论了多重分析技术在生物标志物开发过程中的潜在应用,并强调了对所得数据进行机器学习分析的潜在挑战。