Prediction of hospital mortality in intensive care unit patients from clinical and laboratory data: A machine learning approach.

BACKGROUND: Intensive care unit (ICU) patients demand continuous monitoring of several clinical and laboratory parameters that directly influence their medical progress and the staff's decision-making. Those data are vital in the assistance of these patients, being already used by several scoring systems. In this context, machine learning approaches have been used for medical predictions based on clinical data, which includes patient outcomes. AIM: To develop a binary classifier for the outcome of death in ICU patients based on clinical and laboratory parameters, a set formed by 1087 instances and 50 variables from ICU patients admitted to the emergency department was obtained in the "WiDS (Women in Data Science) Datathon 2020: ICU Mortality Prediction" dataset. METHODS: For categorical variables, frequencies and risk ratios were calculated. Numerical variables were computed as means and standard deviations and Mann-Whitney U tests were performed. We then divided the data into a training (80%) and test (20%) set. The training set was used to train a predictive model based on the Random Forest algorithm and the test set was used to evaluate the predictive effectiveness of the model. RESULTS: A statistically significant association was identified between need for intubation, as well predominant systemic cardiovascular involvement, and hospital death. A number of the numerical variables analyzed (for instance Glasgow Coma Score punctuations, mean arterial pressure, temperature, pH, and lactate, creatinine, albumin and bilirubin values) were also significantly associated with death outcome. The proposed binary Random Forest classifier obtained on the test set (n = 218) had an accuracy of 80.28%, sensitivity of 81.82%, specificity of 79.43%, positive predictive value of 73.26%, negative predictive value of 84.85%, F1 score of 0.74, and area under the curve score of 0.85. The predictive variables of the greatest importance were the maximum and minimum lactate values, adding up to a predictive importance of 15.54%. CONCLUSION: We demonstrated the efficacy of a Random Forest machine learning algorithm for handling clinical and laboratory data from patients under intensive monitoring. Therefore, we endorse the emerging notion that machine learning has great potential to provide us support to critically question existing methodologies, allowing improvements that reduce mortality.

Simple and efficient screening of patients with Fabry disease with high resolution melting.

BACKGROUND: Fabry disease (FD [MIM: 301500]) is a disorder caused by mutations in the alpha-galactosidase gene (GLA), which presents great allelic heterogeneity. The development of fast screening methods may reduce costs and length of diagnosis, being particularly important for screening programs of high-risk female patients. Therefore, the purpose of this study was to develop a pre-sequencing genetic screening method based on high resolution melting (HRM) analysis. METHODS: We performed HRM analysis in one hundred and three individuals, 79 females and 24 males, with a total of 27 different variants in 30 different genotypes. We standardized a protocol using EvaGreen, a release-on-demand dye specific for HRM, added to the PCR reaction. Amplification was performed in a conventional real-time system with HRM capability. RESULTS: All genotypes in all amplicons were distinguishable from wild type. In most amplicons it was even possible to differentiate each genotype from the others. CONCLUSION: We developed a simple, fast and highly sensitive HRM based protocol that may facilitate genetic screening of FD.

Trk Inhibition Reduces Tumorsphere Formation and Changes Expression of Stemness Markers in SK-ES-1 Ewing Sarcoma Cells / Inibição de Trk Reduz a Formação de Tumoresferas e Modifica a Expressão de Marcadores Tronco Tumorais em Células de Sarcoma de Ewing SK-ES-1 / Inhibición de Trk Reduce la Formación de Esferoides Tumorales y Cambia la Expresión de Marcadores de Stemness en Células de Sarcoma de Ewing SK-ES-1

Introduction: Ewing sarcoma (ES) is a highly aggressive type of childhood cancer characterized by a chromosomal translocation resulting in fusions between the gene encoding EWS RNA Binding Protein 1 (EWSR1) and one gene of the ETS family, most frequently FLI-1, resulting in the EWS-FLI1 aberrant transcription factor. ES tumors can contain a subpopulation of cells showing cancer stem cell (CSC) features, which express stemness markers including CD133, OCT4 (Octamer-binding transcription factor 4), and NANOG, and display capacity to form tumorspheres likely enriched in CSCs. Neurotrophin (NT) receptors of the tropomyosin receptor kinase (Trk) family (TrkA, TrkB, and TrkC) may play a role in stimulating ES progression, but their possible role in CSCs remains unknown. Objective: To verify the effect of Trks inhibition on the formation of tumorspheres as well as the gene expression of stem markers. Method: The cells were dissociated and the formation of spheres was induced with supplemented culture medium and the K252a treatment was performed. After RNA extraction, mRNA expression levels of target genes Prom1 (CD133), OCT4 (POU5F1), SOX2, and Musashi-1 (MSI1) were analyzed by qPCR. Results: The pan-Trk inhibitor K252a (100 or 500 mM) hindered tumorsphere formation in human SK-ES-1 ES cell cultures. K252a also reduced mRNA expression of Prom1 (CD133-coding gene) while enhancing expression of OCT4. No changes in mRNA levels of SOX2 or Musashi-1 were observed. Conclusion: These findings provide the first evidence suggesting that Trk activity can influence stemness in ES cells

Introdução: O sarcoma de Ewing (SE) é um tipo altamente agressivo de câncer infantil caracterizado por uma translocação cromossômica que resulta em fusões entre o gene que codifica a proteína de ligação a RNA EWS 1 (EWSR1) e um gene da família ETS, mais frequentemente o FLI-1, resultando no fator de transcrição aberrante EWS-FLI1. Os tumores de SE podem conter uma subpopulação de células com características de células-tronco tumorais (CTT), que expressam marcadores de pluripotência como CD133, OCT4 e NANOG, e têm a capacidade de formar esferas tumorais provavelmente enriquecidas em CTT. Os receptores de neurotrofinas (NT) da família de receptor de quinase de tropomiosina (Trk) (TrkA, TrkB e TrkC) podem desempenhar um papel no estímulo à progressão do SE, mas seu possível papel nas CTT permanece desconhecido. Objetivo: Verificar o efeito da inibição dos Trk na formação de tumoresferas, bem como na expressão gênica de marcadores de pluripotência. Método: As células foram dissociadas, a formação de esferas com meio de cultura suplementado foi induzida e realizou-se o tratamento com K252a. Após a extração de RNA, os níveis de expressão de mRNA dos genes-alvo Prom1 (CD133), OCT4 (POU5F1), SOX2 e Musashi-1 (MSI1) foram analisados por qPCR. Resultados: O inibidor pan-Trk K252a (100 ou 500 mM) impediu a formação de esferas tumorais em culturas de células de SE humanas SK-ES-1. O K252a também reduziu a expressão de mRNA de Prom1 (o gene que codifica CD133), enquanto aumentou a expressão de OCT4. Não foram observadas mudanças nos níveis de mRNA de SOX2 ou Musashi-1. Conclusão: Essas descobertas fornecem as primeiras evidências, sugerindo que a atividade dos Trk possa influenciar a pluripotência nas células de SE

Introducción: El sarcoma de Ewing (SE) es un tipo de cáncer infantil altamente agresivo caracterizado por una translocación cromosómica que resulta en fusiones entre el gen que codifica la proteína de unión a RNA EWS 1 (EWSR1) y un gen de la familia ETS, más frecuentemente FLI-1, lo que resulta en el factor de transcripción aberrante EWS-FLI1. Los tumores del SE pueden contener una subpoblación de células que presentan características de células madre cancerosas (CMC), las cuales expresan marcadores de pluripotencia como CD133, OCT4 y NANOG, y muestran la capacidad de formar esferas tumorales probablemente enriquecidas en CMC. Los receptores de neurotrofinas (NT) de la familia del receptor de quinasa de tropomiosina (Trk) (TrkA, TrkB y TrkC) podrían desempeñar un papel en el estímulo de la progresión del SE, pero su posible papel en las CMC aún es desconocido. Objetivo: Verificar el efecto de la inhibición de los Trk en la formación de esferoides tumorales, así como en la expresión génica de marcadores de pluripotencia. Método: Las células fueron disociadas e inducidas a formar esferas con un medio de cultivo suplementado y se realizó el tratamiento con K252a. Después de la extracción de ARN, los niveles de expresión de ARNm de los genes objetivo Prom1 (CD133), OCT4 (POU5F1), SOX2 y Musashi-1 (MSI1) se analizaron mediante qPCR. Resultados: El inhibidor pan-Trk K252a (100 o 500 mM) evitó la formación de esferas tumorales en cultivos de células de SE humanas SK-ES-1. El K252a también redujo la expresión de ARNm de Prom1 (el gen que codifica CD133), mientras que aumentaba la expresión de OCT4. No se observaron cambios en los niveles de ARNm de SOX2 o Musashi-1. Conclusión: Estos hallazgos proporcionan las primeras evidencias que sugieren que la actividad de Trk puede influir en la pluripotencia en las células del SE