RESUMO

BACKGROUND: Normalization is a critical step in the analysis of single-cell RNA-sequencing (scRNA-seq) datasets. Its main goal is to make gene counts comparable within and between cells. To do so, normalization methods must account for technical and biological variability. Numerous normalization methods have been developed addressing different sources of dispersion and making specific assumptions about the count data. MAIN BODY: The selection of a normalization method has a direct impact on downstream analysis, for example differential gene expression and cluster identification. Thus, the objective of this review is to guide the reader in making an informed decision on the most appropriate normalization method to use. To this aim, we first give an overview of the different single cell sequencing platforms and methods commonly used including isolation and library preparation protocols. Next, we discuss the inherent sources of variability of scRNA-seq datasets. We describe the categories of normalization methods and include examples of each. We also delineate imputation and batch-effect correction methods. Furthermore, we describe data-driven metrics commonly used to evaluate the performance of normalization methods. We also discuss common scRNA-seq methods and toolkits used for integrated data analysis. CONCLUSIONS: According to the correction performed, normalization methods can be broadly classified as within and between-sample algorithms. Moreover, with respect to the mathematical model used, normalization methods can further be classified into: global scaling methods, generalized linear models, mixed methods, and machine learning-based methods. Each of these methods depict pros and cons and make different statistical assumptions. However, there is no better performing normalization method. Instead, metrics such as silhouette width, K-nearest neighbor batch-effect test, or Highly Variable Genes are recommended to assess the performance of normalization methods.

Assuntos

RESUMO

OBJECTIVES: Because published data about the variability of reticulocyte counts in children are scarce, the interindividual biological variability of the automated reticulocyte count and its maturation fractions according to age and sex were analyzed. METHODS: A retrospective, observational, analytical study was designed to establish and compare normal values of the automated reticulocyte count and its maturation fractions in different age and sex groups. The sample was drawn from results of CBC counts performed in children aged between 2 months and 18 years using an indirect sampling methodology. RESULTS: A total of 9,362 CBC counts were analyzed. Automated reticulocyte count decreased between 2 months and 3 years of age and slowly increased thereafter, showing higher values in girls up to the age of 9 years, and equalized by sex thereafter. Immature reticulocyte fraction increased until 7 months of age; decreased progressively until 4 years of age; and then showed a discreet but constant rise, with significantly higher values in boys older than 1 year. The low-fluorescence fraction was relatively steady, with significantly higher values in girls aged 8 months and older. CONCLUSIONS: The automated reticulocyte count and its maturation fractions show significant variations related to age and sex in pediatric patients.

Assuntos

RESUMO

Los límites analíticos de desempeño (LAD) forman parte del diseño del programa de control de calidad analítico. Los objetivos de este trabajo fueron determinar imprecisión (CV), error sistemático (ES) y error total (ET) de 14 analitos de Química Clínica en los sectores de planta y de guardia del laboratorio del HIGA O. Alende de Mar del Plata, evaluar su desempeño analítico según variabilidad biológica y comparar los datos obtenidos. Se realizó un estudio retrospectivo con el registro de controles de calidad y se utilizaron los LAD derivados de VB para obtener las especificaciones de calidad para CV, ES y ET. Respecto del CV, cumple con los criterios el 79% en el sector planta y el 64% en el de guardia. Y en cuanto al ET, cumple el 90% y el 75%, respectivamente. En conclusión, aunque la mayoría de los analitos evaluados cumplen al menos con los criterios mínimos establecidos, los resultados ponen de manifiesto la necesidad de mejorar el desempeño analítico. Detectar los tipos de errores presentes en el proceso de laboratorio es el primer paso para instaurar controles de procedimiento y análisis, soluciones que permitirán mejorar la calidad analítica, uno de los pilares que optimizan la seguridad del paciente.

The analytical limits of performance (ALP) are part of the programme design of analytical quality control. The goals of this study were to determine imprecision (VC), systematic error (SE) and total error (TE) of 14 clinical chemistry analytes in the routine and emergency laboratory of HIGA O. Alende of Mar del Plata, to evaluate their analytical performance in accordance with biological variability, and compare the data obtained. A retrospective study was performed using the record of quality controls. The ALP were obtained from BV to get the quality specifications for VC, SE and TE. Regarding VC, 78% of the analytes meet the criteria in the routine laboratory and 64% in the emergency. Regarding TE, in the routine laboratory, 90% meet the TE criteria, and 75% in the emergency. It can be concluded that, although most of the evaluated analytes meet at least the minimum established criteria, the results highlight the need to improve the analytical performance. Detecting the types of errors in the laboratory process is the first step to establish procedural and analysis controls. These are solutions that will improve the analytical quality, one of the pillars to optimize patient safety.

Os limites analíticos de desmpeño (LAD) são parte do desenho do programa de controle de qualidade analítico. Os objetivos deste estudo foram determinar a imprecisão (CV), o erro sistemático (ES) e o erro total (ET) de 14 analitos de Química Clínica nas áreas de planta e de plantão do laboratório HIGA O. Alende de Mar del Plata, avaliar seu desempenho analítico de acordo com a variabilidade biológica e comparar os dados obtidos. Um estudo retrospectivo foi realizado com o registro de controles de qualidade utilizando os LAD derivados de VB para obter as especificações de qualidade para CV, ES e ET. Quanto ao CV, 79% cumpre com os critérios no setor da planta e 64% no setor de plantão. E, quanto ao ET, cumpre 90% e 75% respectivamente. Em conclusão, embora a maioria dos analitos testados cumpra pelo menos com os critérios mínimos estabelecidos, os resultados destacam a necessidade de melhorar o desempenho analítico. Detectar os tipos de erros encontrados no processo de laboratório é o primeiro passo para estabelecer controles de procedimento e análise, soluções que permitirão melhorar a qualidade analítica, um dos pilares que otimizam a segurança do paciente.

Assuntos

RESUMO

El objetivo del trabajo fue comparar los requerimientos de calidad (RC) de Variabilidad Biológica (VB) con el Estado Actual de la Metodología (EA) en ocho analitos de hemostasia. Se determinó el EA calculando el Coeficiente de Variación promedio ponderado (CVpp) de al menos 6 evaluaciones externas: RIQAS (ET1) y CAP (ET2). Los datos de Error Total aceptable (ETa) por VB mínimo (VBm) y deseable (VBd) se calcularon a partir de los CV intra e inter individuos reportados en www.westgard.com. Los datos obtenidos: Tiempo de Protrombina (TP segundos): ETVBm 7,9%, ETVBd 5,3%, ET1 19%, ET2 13%; Tiempo parcial de tromboplastina activada: (APTT segundos): ETVBm 6,7%, ETVBd 4,5%, ET1 23%, ET2 11%. INR: ETVBm 7,9%, ETVBd 5,3%, ET1 20%, ET2 16%; Fibrinógeno: ETVBm 20,4%, ETVBd13,6%, ET 10%, ET2 16%, FVIII: ETVBm13,3%, ETVBd 8,9%, ET1 30%, ET2 45%, FVII ETVBm16,1%, ETVBd 10,7%, ET1 31%, ET2 42%, Proteína C cromogénica (PCc) ETVBm 28%, ETVBd 18,7%, ET1 36%, ET2 25%; Proteína S libre (PSl ): ETVBm 31,1%, ETVBd 20,7%, ET1 18%, ET2 28%; Antitrombina cromogénica (ATc): ETVBm 12,5%, ETVBd 8,9%, ET1 18%, ET2 28%. Los únicos analitos que cumplen con el requerimiento de calidad de VBm o VBd son: fibrinógeno, PC y PS. Si bien cada laboratorio puede decidir las especificaciones de calidad que desea aplicar, la cuestión a debatir es: "cuál es el requerimiento de calidad deseable para la utilidad clínica de estos ensayos".

The aim of this work was to compare the quality requirements of biological variability (BV) with the state of the art (SA) in eight hemostasis analytes. SA was determined by calculating the weighted average coefficient of vari ation (CVwa) of at least 6 external evaluations: RIQAS (ET1) and CAP (ET2). Data acceptable total error (TEa) for minimum and desirable biological variability (VBm y VBd) was calculated from the coefficient of variation (CV) within-subject and between subject www.westgard.com reported. The following was the data : Prothrombin time ( PT second): ETVBm 7.9%, ETVBd 5.3%, ET1 19%, ET2 13%; Activated partial thromboplastin time (second APTT): ETVBm 6.7%, ETVBd 4.5%, ET1 23%, ET2 11%; INR: ETVBm 7.9%, ETVBd 5.3%, ET1 20%, ET2 16%; Fibrinogen: ETVBm 20.4% ETVBd 13.6% ET1 20%, ET2 16%, FVIII: ETVBm 13.3%, ETVBd 8.9%, ET1 30%, ET2 45% ; FVII: ETVBm 16.1%, ETVBd 10.7%, ET1 31%, ET2 42%; chromogenic Protein C (PCc): ETVBm 28%, ETVBd 18.7%, ET1 36%, ET2 25%; free Protein S (PSf ): ETVBm 31.1% ETVBd 20.7%, ET1 18%, ET2 28%; chromogenic Antithrombin (ATc): ETVBm 12.5%, ETVBd8.9%, ET1 18%, ET2 28%.The only analytes that meet the VBm or VBd quality requirement are fibrinogen, PC and PS. While each laboratory can decide the quality specifications it wants to apply, the issue to be discussed is: "what is the desirable quality requirement for clinical usefulness of these tests?"

O objetivo do trabalho foi comparar os requisitos de qualidade (RQ) de variabilidade biológica (VB) com o estado atual da metodologia (EA) em oito analitos de hemostasia. Foi determinada a EA através do cálculo do coeficiente de variação médio ponderado (CVmp) de pelo menos 6 avaliações externas: RIQAS (ET1) e CAP (ET2). Os dados de erro total admissível (ETa) para VB mínimo desejável (VBm) e (VBd) foram calculados a partir do CV intra e inter indivíduos reportados em www.westgard.com. Os dados obtidos: Tempo de Protrombina (TP segundos) ETVBm 7,9%, ETVBd 5,3%, ET1 19%, ET2 13% ; Tempo parcial de tromboplastina ativada (APTT segundos): ETVBm 6,7%, ETVBd 4,5%, ET1 23%, ET2 11%; INR: ETVBm 7,9%, ETVBd 5,3%, ET1 20%, ET2 16%; Fibrinogênio: ETVBm 20.4%, ETVBd 13,6%, ET1 20%, ET2 16%; FVIII: ETVBm 13,3%, ETVBd 8,9%, ET1 30%, ET2 45%; FVII: ETVBm 16,1%, ETVBd 10,7%, ET1 31%, ET2 42%; Proteína C cromogênica (PCc): ETVBm 28% ETVBd 18,7%, ET1: 36%, ET2: 25%; Proteína S livre (PSl ): ETVBm: 31,1%, ETVBd 20,7%, ET1: 18%, ET2: 28%; Antitrombina cromogênica (ATc): ETVBm12,5%, ETVBd 8.9%, ET1 18%, ET2 28%. Os únicos analitos que atendem o requisito de qualidade de VBm ou VBd são: fibrinogênio, PC e PS. Embora cada laboratório possa decidir as especificações de qualidade que deseja aplicar, a questão a ser discutida é "qual é o requisito de qualidade desejável para a utilidade clínica destes testes?".

Assuntos

RESUMO

El valor de referencia del cambio (VRC) es el valor máximo que es permisible cambie el resultado de un analito entre dos mediciones sucesivas en un mismo paciente, sin que esta diferencia sea de relevancia clínica. Incluye critérios basados en la variabilidad biológica intraindividual (CVI) y la imprecisión analítica (CVA). La principal utilidad de los marcadores tumorales (MT) es el monitoreo de pacientes, resultando más apropiado informar el VRC que evaluar un resultado con su valor de referencia, como lo indica su bajo índice de individualidad. El objetivo fue evaluar la utilidad del VRC para detectar un cambio significativo entre resultados sucesivos en los principales MT. Se analizaron datos de sueros de controles de calidad de MT desde mayo de 2010 a febrero de 2014, se calculó el CVA%, y los datos de CVI % fueron obtenidos de bibliografía. Se calcularon los VRC para cada MT. Para los MT evaluados: AFP, CEA, CA125, CA15-3, CA19-9, PSA y tiroglobulina, los VRC fueron: 29.7, 32.3, 58.0, 16.3, 38.3, 42.7 y 34.2% respectivamente (p<0.05). Estos valores se compararon con datos bibliográficos. El VRC es un dato útil para el médico ya que colabora en la correcta interpretación de resultados seriados durante el seguimiento de pacientes, en la evaluación del tratamiento o en la estimación de recurrencias. Le permite saber si la diferencia encontrada entre dos valores consecutivos representa un cambio em el estado de salud del paciente. Nuestros VRC resultaron comparables con los de literatura

The reference change value (RCV) is the maximum allowable change between two consecutive results with no meaningful clinical relevance. It is analyzed within individual biological variability (CVI ) and analytical imprecision (CVA) criteria. For tumor markers (TM) monitoring is more appropriate to report RCV than reference interval due to their low individuality index. The aim of the study was to evaluate the usefulness of RCV to indicate a significant change between two consecutives TM results. Data from MT quality control serums (QC) were analyzed from May 2010 to February 2014, the imprecision was calculated as CVA% and CVI % data was obtained from literature. The RCV for each MT was calculated. The RCV for AFP, CEA, CA125, CA15-3, CA19-9, PSA and thyroglobulin were 29.7, 32.3, 58.0, 16.3, 38.3, 42.7 and 34.2% respectively (p < 0.05). These values were compared with literature data. The RCV is an appropriate tool for the clinicians and aids for the correct interpretation of results in the monitoring of patients, in treatment evaluation or estimation of recurrence. Physicians can determine whether the differences found between two successive values represent a change in the health status of the patient. The RCV calculated were comparable with those obtained in literature

Assuntos

RESUMO

Las determinaciones realizadas en laboratorios clínicos deben considerar las variaciones propias de muestras biológicas sólo por ser parte de un sistema complejo. El presente trabajo estudió el efecto del ayuno y hora (h) de toma de muestra en determinaciones de glucosa, urea, creatinina, colesterol, triglicéridos, HDl-colesterol, LDL-colesterol, proteínas, albúmina y ácido úrico. Se conformó una muestra con 31 personas de Caracas con edades entre 18 y 54 años. A cada una se le tomó dos muestras de sangre: una a las 7am (ayuno 12H) y otra (diferente día) a la 1 pm (ayuno 6h, desayuno 7am estandarizado). Las mustras se procesaron en Konelab-20. Se comprobó la normalidad de las distribuciones y se compararon los grupos según el ayuno realizado, todo bajo técnicas de estadística inferencial en SPSS 12.0. Según los resultados no se encontraron diferencias significativas (p>0,05) en las muestras tomadas con ayuno de 12h y 6h en Glucosa, Proteínas totales, albúmina, ácido úrico, colesterol, triglicéridos, HDL-colesterol y LDL-colesterol, no así para el caso de creatinina y urea, donde se observaron diferencias significativas (p<0,05). No existen variaciones de los resultados en un mismo paciente, con dos muestras tomadas a diferentes horas, una en la mañana con 12h de ayuno y otra, en la tarde con 6h de ayuno para el colesterol, triglicéridos, fracciones lípidicas, proteínas, albúmina, acido úrico y glucosa. En el caso de urea y creatinina, existen diferencias significativas, por lo que se recomienda estableces intervalos de referencia biológicos para las determinaciones realizadas con 6h de ayuno.

The determination realized in clinical laboratories must considerer the individual variations of biological samples, only for being a part of a complex system. The present work studied the effect the fasting and hour (h) of capture of samples in determinations of glucose, urea, creatinina, cholesterol, triglicéridos, HDl-cholesterol, LDL-cholesterol, proteins, albumen and uric acid. We studied a sample of 31 persons from Caracas with ages between 18 and 54 years. To each one was drawn two samples of blood: one at 7am (12h fasting) and the other a different day at 1pm (6h fastinf, post standarized breakfast at 7am). The samples were tested in a Konelab-20. The normality of the distributions was verified and the groups and were compared according to the fasting under test of statics inferencial in SPSS 12.0. No significant differences found (p>0,05) in the samples taken with fasting of 12h and 6h in HDL, Glucose, Total Proteins, albumen, uric acid, cholesterol, triglicéridos, cholesterol and LDL-cholesterol, this was not the case for creatinina and urea, where significant differences were observed (p<0,05). Variation of the results does not exist in the same patient, with two samples taken at different hours, one in the morning with 12 h of fasting and other one, in the evening with 6h of fasting for the cholesterol, triglicéridos, lipidic fractions, proteins, albumen, uric and glucose. In case of urea and creatinina, significant difference exist, so we recommend establishing reference biological intervals for the determinations of these components realized with 6h of fasting.

Assuntos