Narváez Bandera, Isis Y.

Profile Picture

Filters

2017 - 201912020 - 20221
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Publication
    Biooptimatics studies to further substantiate a liaison between Flaviviridae infections and Parkinson´s disease
    (2022-12-13) Narváez Bandera, Isis Y.; Isaza Brando, Clara E.; College of Engineering; Cabrera Ríos, Mauricio; González Méndez, Ricardo; Latorre Esteves, Magda; Other; Valderrama Fuquen, Clara
    The theory of a viral etiology for Parkinson's disease (PD) has gained growing evidence over the past decade. This work presents an optimization-based meta-analysis to investigate whether Flaviviridae infections, such as Dengue Virus (DV) and Hepatitis C Virus (HCV), are risk factors for PD. To this end, 25 publicly available case-control studies for DV, HCV, and PD were analyzed. To handle this amount of data, it was first necessary to code, automate and enhance the capabilities of the BioOptimatics methods previously proposed by our research group. The optimization-base methods include Multiple criteria optimization (MCO), which selects genes with the largest expression changes, and the Minimum Spanning Tree (MST), which proposes a maximum correlated structure. Therefore, the present work describes the development of an open-source tool in RStudio to enable: (1) individual analysis of single datasets, (2) analysis by sex, and (3) meta-analysis with up to five datasets. The capabilities afforded by the code include license-free portability and the possibility to conduct analyses via modest computer hardware, such as personal laptops. The tool provides affordable, repeatable, and objective detection of differentially expressed genes and potential signaling pathways from microarrays and RNA-sequence datasets. Making use of these advantages, this study identified over 17 genes (IFI27, CFD, ARHGDIB, CSTA, UBA52, RPL9, HBD, RPL39, MMP9, TUBB2A, RPS16, RPS11, RPL18, SLC4A1, RPL30, CD52, and FAU) that could be important in developing PD after DV infection, and over 19 genes (MT1H, MYOM2, RPL18, S100A12, IFIT1, KRT23, GPX3, SRGN, PGK1, SLC30A2, SAA1, SERPI1, FOS, GRN, LYZ, CEBPD, EPB41L3, GSTM1, and CXCL1) that could be a link between HCV and PD. Finally, we propose several possible pathways involving ubiquitination, inflammation, and neurodegeneration that could be crucial in developing PD after DV or HCV infection. Further examination of the link between flavivirus infection and PD is warranted.
  • Thumbnail Image
    Publication
    An integrative data-driven approach to identify molecular patterns in breast cancer patients
    (2017) Narváez Bandera, Isis Y.; Torres García, Wandaliz; College of Engineering; Cabrera Ríos, Mauricio; Domenech García, Maribella; Department of Industrial Engineering; Rodríguez Román, Daniel
    El cáncer de mama es una enfermedad heterogénea del genoma que necesita mejores diagnósticos y tratamientos a través de la caracterización de patrones genómicos e interacciones. Actualmente, su comprensión molecular es aún insuficiente incluso con los avances en las tecnologías genómicas. Por lo tanto, esta tesis presenta un enfoque de minería de datos en varias etapas para discriminar los subtipos de cáncer de mama a través de la integración de datos altamente dimensionales de diferentes plataformas genómicas utilizando técnicas de selección y clasificación de características. Esta metodología nos permitió extraer patrones que desempeñan un papel crítico en la clasificación de los subtipos de cáncer de mama (es decir, la subexpresión de FOXA1 para basal). Además, esta tesis proporciona una nueva métrica capaz de evaluar y clasificar las interacciones entre las características pertinentes utilizando un criterio de prevalencia y el clasificador Random Forest. Esta métrica identificó una lista de interacciones de variables importantes para discriminar subtipos. Entre las principales interacciones, encontramos un conjunto de genes correlacionados interactuando frecuentemente con FOXA1 o MLPH tales como CEP55 y UBET2.