Sonia L Mestizo Gutiérrez1, Marisol Herrera Rivero2, Nicandro Cruz Ramírez3, Elena Hernández4, Gonzalo E Aranda-Abreu5. 1. Doctorado en Investigaciones Cerebrales, Universidad Veracruzana, Av. Luis Castelazo Ayala S/N, Xalapa, Veracruz 91190, Mexico. 2. Doctorado en Ciencias Biomédicas, Universidad Veracruzana, Av. Luis Castelazo Ayala S/N, Xalapa, Veracruz, Mexico. 3. Departamento de Inteligencia Artificial, Universidad Veracruzana, Sebastián Camacho 5, Centro, Xalapa, Veracruz 91000, Mexico. 4. Centro de Investigaciones Cerebrales, Cuerpo Académico de Neuroquímica, Universidad Veracruzana, Av. Luis Castelazo Ayala S/N, Xalapa, Veracruz, Mexico. 5. Centro de Investigaciones Cerebrales, Cuerpo Académico de Neuroquímica, Universidad Veracruzana, Av. Luis Castelazo Ayala S/N, Xalapa, Veracruz, Mexico. Electronic address: garanda@uv.mx.
Abstract
BACKGROUND: Alzheimer's disease (AD) is characterized by a gradual loss of memory, orientation, judgement and language. There is still no cure for this disorder. AD pathogenesis remains fairly unknown and its underlying molecular mechanisms are not yet fully understood. Several studies have shown that the abnormal accumulation of beta-amyloid and tau proteins occurs 10 to 20 years before the onset of symptoms of the disease, so it is extremely important to identify changes in the brain before the first symptoms. METHODS: We used decision trees to classify 31 individuals (9 healthy controls and 22 AD patients in three different stages of disease) according to the expression of 69 genes previously reported in a meta-analysis, plus the expression levels of APP, APOE, BACE1, NCSTN, PSEN1, PSEN2 and MAPT. We also included in our analysis the MMSE (Mini-Mental State Examination) scores and number of NFT (neurofibrillary tangles). RESULTS: Results allowed us to generate a model of classification values for different AD stages of severity, according to MMSE scores, and achieve the identification of the expression level of protein tau that may possibly determine the onset (incipient stage) of AD. DISCUSSION: We used decision trees to model the different stages of AD (severe, moderate, incipient and control) based on the meta-analysis of gene expression levels plus MMSE and NFT scores. Both classifiers reported the variable MMSE as most informative, however it we were found that the protein tau also an important role in the onset of AD.
BACKGROUND:Alzheimer's disease (AD) is characterized by a gradual loss of memory, orientation, judgement and language. There is still no cure for this disorder. AD pathogenesis remains fairly unknown and its underlying molecular mechanisms are not yet fully understood. Several studies have shown that the abnormal accumulation of beta-amyloid and tau proteins occurs 10 to 20 years before the onset of symptoms of the disease, so it is extremely important to identify changes in the brain before the first symptoms. METHODS: We used decision trees to classify 31 individuals (9 healthy controls and 22 ADpatients in three different stages of disease) according to the expression of 69 genes previously reported in a meta-analysis, plus the expression levels of APP, APOE, BACE1, NCSTN, PSEN1, PSEN2 and MAPT. We also included in our analysis the MMSE (Mini-Mental State Examination) scores and number of NFT (neurofibrillary tangles). RESULTS: Results allowed us to generate a model of classification values for different AD stages of severity, according to MMSE scores, and achieve the identification of the expression level of protein tau that may possibly determine the onset (incipient stage) of AD. DISCUSSION: We used decision trees to model the different stages of AD (severe, moderate, incipient and control) based on the meta-analysis of gene expression levels plus MMSE and NFT scores. Both classifiers reported the variable MMSE as most informative, however it we were found that the protein tau also an important role in the onset of AD.