RESUMEN
iMole is a platform that automatically extracts images and captions from biomedical literature. Images are tagged with terms contained in figure captions by means of a sophisticate text-mining tool. Moreover, iMole allows the user to upload directly their own images within the database and manually tag images by curated dictionary. Using iMole the researchers can develop a proper biomedical image database, storing the images extracted from paper of interest, image found on the web repositories, and their own experimental images. In order to show the functioning of the platform, we used iMole to build a 2DE database. Briefly, tagged 2DE gel images were collected and stored in a searchable 2DE gel database, available to users through an interactive web interface. Images were obtained by automatically parsing 16,608 proteomic publications, which yielded more than 16,500 images. The database can be further expanded by users with images of interest trough a manual uploading process. iMole is available with a preloaded set of 2DE gel data at http://imole.biodigitalvalley.com.
Asunto(s)
Minería de Datos/métodos , Sistemas de Administración de Bases de Datos , Bases de Datos Factuales , Internet , Programas Informáticos , Electroforesis en Gel Bidimensional , Procesamiento de Imagen Asistido por Computador , Almacenamiento y Recuperación de la InformaciónRESUMEN
Diagnosis of Parkinson's disease (PD) is currently assessed by the clinical evaluation of extrapyramidal signs. The identification of specific biomarkers would be advisable, however most studies stop at the discovery phase, with no biomarkers reaching clinical exploitation. To this purpose, we developed an automated literature analysis procedure to retrieve all the background knowledge available in public databases. The bioinformatic platform allowed us to analyze more than 51,000 scientific papers dealing with PD, containing information on 4121 proteins. Out of these, we could track back 35 PD-related proteins as present in at least two published 2-DE maps of human plasma. Then, 9 different proteins (haptoglobin, transthyretin, apolipoprotein A-1, serum amyloid P component, apolipoprotein E, complement factor H, fibrinogen γ, thrombin, complement C3) split into 32 spots were identified as a potential diagnostic pattern. Eventually, we compared the collected literature data to experimental gels from 90 subjects (45 PD patients, 45 non-neurodegenerative control subjects) to experimentally verify their potential as plasma biomarkers of PD.
Asunto(s)
Proteínas Sanguíneas/metabolismo , Minería de Datos , Procesamiento Automatizado de Datos , Enfermedad de Parkinson/sangre , Anciano , Biomarcadores/sangre , Femenino , Humanos , Masculino , Persona de Mediana Edad , Enfermedad de Parkinson/diagnósticoRESUMEN
MOTIVATION: The two-dimensional electrophoresis (2-DE) pattern of proteins is thought to be specifically related to the physiological or pathological condition at the moment of sample preparation. On this ground, most proteomic studies move to identify specific hallmarks for a number of different conditions. However, the information arising from these investigations is often incomplete due to inherent limitations of the technique, to extensive protein post-translational modifications and sometimes to the paucity of available samples. The meta-analysis of proteomic data can provide valuable information pertinent to various biological processes that otherwise remains hidden. RESULTS: Here, we show a meta-analysis of the PD protein DJ-1 in heterogeneous 2-DE experiments. The protein was shown to segregate into specific clusters associated with defined conditions. Interestingly, the DJ-1 pool from neural tissues displayed a specific and characteristic molecular weight and isoelectric point pattern. Moreover, changes in this pattern have been related to neurodegenerative processes and aging. These results were experimentally validated on human brain specimens from control subjects and PD patients. AVAILABILITY: ImageJ is a public domain image processing program developed by the National Institutes of Health and is freely available at http://rsbweb.nih.gov/ij. All the ImageJ macros used in this study are available as supplementary material and upon request at info@biodigitalvalley.com. XLSTAT can be purchased online at http://www.xlstat.com/en/home/ at a current cost of approximately 300 EUR.