Expert-level detection of M-proteins in serum protein electrophoresis using machine learning.

OBJECTIVES: Serum protein electrophoresis (SPE) in combination with immunotyping (IMT) is the diagnostic standard for detecting monoclonal proteins (M-proteins). However, interpretation of SPE and IMT is weakly standardized, time consuming and investigator dependent. Here, we present five machine learning (ML) approaches for automated detection of M-proteins on SPE on an unprecedented large and well-curated data set and compare the performance with that of laboratory experts. METHODS: SPE and IMT were performed in serum samples from 69,722 individuals from Norway. IMT results were used to label the samples as M-protein present (positive, n=4,273) or absent (negative n=65,449). Four feature-based ML algorithms and one convolutional neural network (CNN) were trained on 68,722 randomly selected SPE patterns to detect M-proteins. Algorithm performance was compared to that of an expert group of clinical pathologists and laboratory technicians (n=10) on a test set of 1,000 samples. RESULTS: The random forest classifier showed the best performance (F1-Score 93.2 %, accuracy 99.1 %, sensitivity 89.9 %, specificity 99.8 %, positive predictive value 96.9 %, negative predictive value 99.3 %) and outperformed the experts (F1-Score 61.2 ± 16.0 %, accuracy 89.2 ± 10.2 %, sensitivity 94.3 ± 2.8 %, specificity 88.9 ± 10.9 %, positive predictive value 47.3 ± 16.2 %, negative predictive value 99.5 ± 0.2 %) on the test set. Interestingly the performance of the RFC saturated, the CNN performance increased steadily within our training set (n=68,722). CONCLUSIONS: Feature-based ML systems are capable of automated detection of M-proteins on SPE beyond expert-level and show potential for use in the clinical laboratory.

Classical meets malignant hematology: a case of acquired εγδß-thalassemia in clonal hematopoiesis.

Hemoglobinopathies including thalassemias are among the most frequent genetic disorders worldwide. Primarily, these entities result from germline variants in the globin gene clusters and their cis-acting regulatory elements, and thus the WHO classifies thalassemias as inherited diseases. Non-inherited disorders of globin chain synthesis mimicking the phenotype of thalassemias have also been described and are referred to as acquired thalassemias. These forms mainly affect the alpha-globin genes and are observed at much lower frequencies...

Correction to: Rapid and reliable detection of α-globin copy number variations by quantitative real-time PCR.

[This corrects the article DOI: 10.1186/2052-1839-14-4.].

Immediate Neutrophil-Variable-T Cell Receptor Host Response in Bacterial Meningitis.

Bacterial meningitis is a life-threatening disease that evokes an intense neutrophil-dominated host response to microbes invading the subarachnoid space. Recent evidence indicates the existence of combinatorial V(D)J immune receptors in neutrophils that are based on the T cell receptor (TCR). Here, we investigated expression of the novel neutrophil TCRαß-based V(D)J receptors in cerebrospinal fluid (CSF) from human patients with acute-phase bacterial meningitis using immunocytochemical, genetic immunoprofiling, cell biological, and mass spectrometric techniques. We find that the human neutrophil combinatorial V(D)J receptors are rapidly induced in CSF neutrophils during the first hours of bacterial meningitis. Immune receptor repertoire diversity is consistently increased in CSF neutrophils relative to circulating neutrophils and phagocytosis of baits directed to the variable immunoreceptor is enhanced in CSF neutrophils during acute-phase meningitis. Our results reveal that a flexible immune response involving neutrophil V(D)J receptors which enhance phagocytosis is immediately initiated at the site of acute bacterial infection.

Ethanol-Gel Sclerotherapy of Venous Malformations: Effectiveness and Safety.

OBJECTIVE: In the treatment of venous malformations, ethanol may be administered in a gelified form to increase local effects and reduce systemic ones. The purpose of this prospective study was to evaluate the efficacy and safety of a commercially available viscous ethanol gel in the treatment of venous malformations. SUBJECTS AND METHODS: Thirty-one patients (mean age, 23.4 years; age range, 6.6-46.5 years) with venous malformations were prospectively scheduled for two ethanol-gel sclerotherapy sessions. Venous malformations were located at the lower extremity (n = 18), the upper extremity (n = 9), and the face (n = 4). Questionnaires to assess pain, clinical examinations, professional photographs, and contrast-enhanced MRI of the venous malformations were performed before and after therapy to measure therapy-induced changes. Two experienced radiologists blinded to the examination date and clinical status compared photographs and MR images before and after treatment. RESULTS: A mean of 4.2 mL of ethanol gel were administered per session. The technical success rate was 100%. Clinical success, defined as improvement or resolution of symptoms, was noted in 81% of patients. Mean pain score decreased, and the difference was statistically significant (3.9 vs 3.1, p = 0.005). In 54 treatment sessions where follow-up was available, four minor complications occurred. Comparison of photographs and MR images before and after treatment showed improvement in 35% and 93% of patients, respectively. CONCLUSION: Ethanol gel is an effective and safe sclerosing agent in the treatment of venous malformations.

Copy number variation in the ATP-binding cassette transporter ABCC6 gene and ABCC6 pseudogenes in patients with pseudoxanthoma elasticum.

Single mutations in the ATP-binding cassette transporter (ABCC6) gene (OMIM 603234) are known to cause the rare autosomal recessive disease pseudoxanthoma elasticum (PXE). Recently, we have found that copy number variations (CNVs) in pseudogenes of the ABCC6 gene are quite common. The aim of this study was to investigate the frequency and possible contribution of CNV in ABCC6 and its pseudogenes in PXE. Genomic DNA from 212 PXE individuals were examined for copy number by pyrosequencing and quantitative polymerase chain reaction (PCR) and compared with healthy individuals. The frequency of PXE individuals with any CNV was higher than in healthy individuals. The majority of variation comprised known and possibly new deletions in the ABCC6 gene and duplications of the ABCC6P1 and ABCC6P2 genes. ABCC6 deletions and ABCC6P2 duplications were not observed in 142 healthy individuals. In conclusion, by pyrosequencing and quantitative PCR, we were able to detect known and possibly new deletions in the ABCC6 gene that may have caused the PXE phenotype. Pyrosequencing may be used in PXE patients who have obtained incomplete genotype from conventional techniques. The frequency of ABCC6P2 pseudogene duplication was more common in PXE patients than healthy individuals and may affect the PXE phenotype.

Interference of common haemoglobin variants with the Tosoh G7 standard mode HbA1c method.

BACKGROUND: HbA1c methods may be prone to interference by the presence of haemoglobin variants. In contrast to the variant mode of the HbA1c method on the Tosoh G7 instrument, the literature lacks investigations of haemoglobin variant interference with the standard mode. The current study sought to investigate whether different haemoglobin variants interfere with the Tosoh G7 standard mode HbA1c method, and whether present haemoglobin variants are identifiable on respective chromatograms. METHODS: Samples routinely analyzed for HbA1c and suspected of having haemoglobin variants (N = 103) were included. HbA1c was measured on a Tosoh G7 in standard mode (Tosoh Corporation, Japan), and on the DCA Vantage (Siemens, Germany). Haemoglobin variants were identified using the VARIANT(™)ß-Thalassemia Short Program (Bio-Rad Laboratories, Hercules, CA, USA) and by DNA sequencing. RESULTS: The Tosoh G7 in standard mode measured significantly lower HbA1c results (between 1.0 and 2.5 percentage points absolute bias corresponding to between 11 and 27 mmol/mol, p < 0.001) in samples in which common haemoglobin variants (HbS, HbC, HbD or HbE) were present (n = 61). No significant difference in HbA1c (0.04 percentage points, p = 0.74) was found between Tosoh G7 standard mode and DCA Vantage in samples in which haemoglobin variants were absent (n = 36). In contrast to HbS and HbD, HbE and HbC trait could be identified on respective chromatograms. CONCLUSION: The presence of common haemoglobin variants results in falsely low HbA1c measurements on the Tosoh G7 in standard mode. HbS and HbD trait are not identifiable on respective haemoglobin chromatograms.

RNA-sequencing analysis of HepG2 cells treated with atorvastatin.

The cholesterol-lowering drug atorvastatin is among the most prescribed drug in the world. Alternative splicing in a number of genes has been reported to be associated with variable statin response. RNA-seq has proven to be a powerful technique for genome-wide splice variant analysis. In the present study, we sought to investigate atorvastatin responsive splice variants in HepG2 cells using RNA-seq analysis to identify novel candidate genes implicated in cholesterol homeostasis and in the statin response. HepG2 cells were treated with 10 µM atorvastatin for 24 hours. RNA-seq and exon array analyses were performed. The validation of selected genes was performed using Taqman gene expression assays. RNA-seq analysis identified 121 genes and 98 specific splice variants, of which four were minor splice variants to be differentially expressed, 11 were genes with potential changes in their splicing patterns (SYCP3, ZNF195, ZNF674, MYD88, WHSC1, KIF16B, ZNF92, AGER, FCHO1, SLC6A12 and AKAP9), and one was a gene (RAP1GAP) with differential promoter usage. The IL21R transcript was detected to be differentially expressed via RNA-seq and RT-qPCR, but not in the exon array. In conclusion, several novel candidate genes that are affected by atorvastatin treatment were identified in this study. Further studies are needed to determine the biological significance of the atorvastatin responsive splice variants that have been uniquely identified using RNA-seq.

Serum bilirubin concentration in healthy adult North-Europeans is strictly controlled by the UGT1A1 TA-repeat variants.

The major enzyme responsible for the glucuronidation of bilirubin is the uridine 5'-diphosphoglucose glucuronosyltransferase A1 (UGT1A1) enzyme, and genetic variation in the UGT1A1 gene is reported to influence the bilirubin concentration in the blood. In this study, we have investigated which gene-/haplotype variants may be useful for genetic testing of Gilbert's syndrome. Two groups of samples based on serum bilirubin concentrations were obtained from the Nordic Reference Interval Project Bio-bank and Database (NOBIDA): the 150 individuals with the highest bilirubin (>17.5 µmol/L) and the 150 individuals with normal bilirubin concentrations (<17.5 µmol/L). The individuals were examined for the TA6>TA7 variant in the UGT1A1 promoter and 7 tag-SNPs in an extended promoter region of UGT1A1 (haplotype analysis) and in selected SNPs in candidate genes (SLCO1B3, ABCC2 and NUP153). We found significant odds ratios for high bilirubin level for all the selected UGT1A1 variants. However, in stepwise multivariate logistic regression analysis of all genetic variants together with age, sex, country of origin and fasting time, the repeat variants of UGT1A1 TA6>TA7 and SLCO1B3 rs2117032 T>C were the only variants significantly associated with higher bilirubin concentrations. Most individuals with high bilirubin levels were homozygous for the TA7-repeat (74%) while only 3% were homozygous for the TA7-repeat in individuals with normal bilirubin levels. Among individuals heterozygous for the TA7-repeat, a low frequent UGT1A1-diplotype harboring the rs7564935 G-variant was associated with higher bilirubin levels. In conclusion, our results demonstrate that in testing for Gilbert's syndrome, analyzing for the homozygous TA7/TA7-genotype would be appropriate.

Rapid and reliable detection of α-globin copy number variations by quantitative real-time PCR.

BACKGROUND: Alpha-thalassemia is the most common human genetic disease worldwide. Copy number variations in the form of deletions of α-globin genes lead to α-thalassemia while duplications of α-globin genes can cause a severe phenotype in ß-thalassemia carriers due to accentuation of globin chain imbalance. It is important to have simple and reliable methods to identify unknown or rare deletions and duplications in cases in which thalassemia is suspected but cannot be confirmed by multiplex gap-PCR. Here we describe a copy number variation assay to detect deletions and duplications in the α-globin gene cluster (HBA-CNV). RESULTS: Quantitative real-time PCR was performed using four TaqMan® assays which specifically amplify target sequences representing both the α-globin genes, the -α3.7 deletion and the HS-40 region. The copy number for each target was determined by the 2-ΔΔCq method. To validate our method, we compared the HBA-CNV method with traditional gap-PCR in 108 samples from patients referred to our laboratory for hemoglobinopathy evaluation. To determine the robustness of the four assays, we analyzed samples with and without deletions diluted to obtain different DNA concentrations. The HBA-CNV method identified the correct copy numbers in all 108 samples. All four assays showed the correct copy number within a wide range of DNA concentrations (3.2-100 ng/µL), showing that it is a robust and reliable method. By using the method in routine diagnostics of hemoglobinopathies we have also identified several deletions and duplications that are not detected with conventional gap-PCR. CONCLUSIONS: HBA-CNV is able to detect all known large deletions and duplications affecting the α-globin genes, providing a flexible and simple workflow with rapid and reliable results.

Hb Ullevaal [ß78(EF2)Leu→Val; HBB: c.235C>G], a new hemoglobin variant interfering with Hb A1c measurement using a cation exchange high performance liquid chromatography method.

A new hemoglobin (Hb) variant was detected during Hb A1c measurement. DNA sequencing showed heterozygosity for the single nucleotide substitution (C > G) on the ß-globin gene causing an amino acid change [ß78(EF2)Leu→Val; HBB: c.235C > G]. The new Hb variant was designated Hb Ullevaal after the hospital at which it was discovered. Heterozygosity for Hb Ullevaal appears to have no clinical significance. However, it interferes with Hb A1c measurement by cation exchange high performance liquid chromatography (HPLC), causing falsely low Hb A1c concentration when using the Tosoh G7 apparatus in variant mode.

UGT1A1*28 is associated with decreased systemic exposure of atorvastatin lactone.

BACKGROUND: Atorvastatin is commonly used to reduce cholesterol. Atorvastatin acid is converted to its corresponding lactone form spontaneously or via glucuronidation mediated by uridine diphosphate glucuronosyltransferase (UGT) 1A1 and 1A3. Atorvastatin lactone is pharmacologically inactive, but is suspected to be muscle toxic and cause statin-induced myopathy (SIM). A several fold increase in systemic exposure of atorvastatin lactone has previously been observed in patients with SIM compared with healthy control subjects. In this study we aimed to investigate the association between polymorphisms in the UGT1A gene locus and plasma atorvastatin lactone levels. METHODS: DNA was extracted from whole blood obtained from a previous pharmacokinetic study of patients carefully diagnosed as having true SIM (n = 13) and healthy control subjects (n = 15). The UGT1A1*28(TA) 7 , UGT1A3*2, UGT1A3*3, and UGT1A3*6 polymorphisms were detected by pyrosequencing. RESULTS: Carriers of the low-expression allele UGT1A1*28(TA) 7 tended to have lower levels of atorvastatin lactone (p < 0.05) than carriers with the normal-activity allele (TA) 6 . CONCLUSION: The low-expression UGT1A1*28(TA) 7 allele seems to be associated with decreased systemic exposure of the suspected muscle-toxic metabolite atorvastatin lactone.

A novel 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) splice variant with an alternative exon 1 potentially encoding an extended N-terminus.

BACKGROUND: The major rate-limiting enzyme for de novo cholesterol synthesis is 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR). HMGCR is sterically inhibited by statins, the most commonly prescribed drugs for the prevention of cardiovascular events. Alternative splicing of HMGCR has been implicated in the control of cholesterol homeostasis. The aim of this study was to identify novel alternatively spliced variants of HMGCR with potential physiological importance. RESULTS: Bioinformatic analyses predicted three novel HMGCR transcripts containing an alternative exon 1 (HMGCR-1b, -1c, -1d) compared with the canonical transcript (HMGCR-1a). The open reading frame of the HMGCR-1b transcript potentially encodes 20 additional amino acids at the N-terminus, compared with HMGCR-1a. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) was used to examine the mRNA levels of HMGCR in different tissues; HMGCR-1a was the most highly expressed variant in most tissues, with the exception of the skin, esophagus, and uterine cervix, in which HMGCR-1b was the most highly expressed transcript. Atorvastatin treatment of HepG2 cells resulted in increased HMGCR-1b mRNA levels, but unaltered proximal promoter activity compared to untreated cells. In contrast, HMGCR-1c showed a more restricted transcription pattern, but was also induced by atorvastatin treatment. CONCLUSIONS: The gene encoding HMGCR uses alternative, mutually exclusive exon 1 sequences. This contributes to an increased complexity of HMGCR transcripts. Further studies are needed to investigate whether HMGCR splice variants identified in this study are physiologically functional.

Copy number variations of the ATP-binding cassette transporter ABCC6 gene and its pseudogenes.

BACKGROUND: The ATP-binding cassette transporter ABCC6 gene is located on chromosome 16 between its two pseudogenes (ABCC6P1 and ABCC6P2). Previously, we have shown that ABCC6P1 is transcribed and affects ABCC6 at the transcriptional level. In this study we aimed to determine copy number variations of ABCC6, ABCC6P1 and ABCC6P2 in different populations. Moreover, we sought to study the transcription pattern of ABCC6 and ABCC6 pseudogenes in 39 different human tissues. FINDINGS: Genomic DNA from healthy individuals from five populations, Chinese (n = 24), Middle East (n = 20), Mexicans (n = 24), Caucasians (n = 50) and Africans (n = 24), were examined for copy number variations of ABCC6 and its pseudogenes by pyrosequencing and quantitative PCR. Copy number variation of ABCC6 was very rare (2/142; 1.4%). However, one or three copies of ABCC6P1 were relatively common (3% and 8%, respectively). Only one person had a single copy of ABCC6P2 while none had three copies. In Chinese, deletions or duplications of ABCC6P1 were more frequent than in any other population (9/24; 37.5%). The transcription pattern of ABCC6P2 was highly similar to ABCC6 and ABCC6P1, with highest transcription in liver and kidney. Interestingly, the total transcription level of pseudogenes, ABCC6P1 + ABCC6P2, was higher than ABCC6 in most tissues, including liver and kidney. CONCLUSIONS: Copy number variations of the ABCC6 pseudogenes are quite common, especially in populations of Chinese ancestry. The expression pattern of ABCC6P2 in 39 human tissues was highly similar to that of ABCC6 and ABCC6P1 suggesting similar regulatory mechanisms for ABCC6 and its pseudogenes.

A-Subclass ATP-Binding Cassette Proteins in Brain Lipid Homeostasis and Neurodegeneration.

The A-subclass of ATP-binding cassette (ABC) transporters comprises 12 structurally related members of the evolutionarily highly conserved superfamily of ABC transporters. ABCA transporters represent a subgroup of "full-size" multispan transporters of which several members have been shown to mediate the transport of a variety of physiologic lipid compounds across membrane barriers. The importance of ABCA transporters in human disease is documented by the observations that so far four members of this protein family (ABCA1, ABCA3, ABCA4, ABCA12) have been causatively linked to monogenetic disorders including familial high-density lipoprotein deficiency, neonatal surfactant deficiency, degenerative retinopathies, and congenital keratinization disorders. Recent research also point to a significant contribution of several A-subfamily ABC transporters to neurodegenerative diseases, in particular Alzheimer's disease (AD). This review will give a summary of our current knowledge of the A-subclass of ABC transporters with a special focus on brain lipid homeostasis and their involvement in AD.

Metabolic changes in urine during and after pregnancy in a large, multiethnic population-based cohort study of gestational diabetes.

This study aims to identify novel markers for gestational diabetes (GDM) in the biochemical profile of maternal urine using NMR metabolomics. It also catalogs the general effects of pregnancy and delivery on the urine profile. Urine samples were collected at three time points (visit V1: gestational week 8-20; V2: week 28±2; V3 10-16 weeks post partum) from participants in the STORK Groruddalen program, a prospective, multiethnic cohort study of 823 healthy, pregnant women in Oslo, Norway, and analyzed using (1)H-NMR spectroscopy. Metabolites were identified and quantified where possible. PCA, PLS-DA and univariate statistics were applied and found substantial differences between the time points, dominated by a steady increase of urinary lactose concentrations, and an increase during pregnancy and subsequent dramatic reduction of several unidentified NMR signals between 0.5 and 1.1 ppm. Multivariate methods could not reliably identify GDM cases based on the WHO or graded criteria based on IADPSG definitions, indicating that the pattern of urinary metabolites above micromolar concentrations is not influenced strongly and consistently enough by the disease. However, univariate analysis suggests elevated mean citrate concentrations with increasing hyperglycemia. Multivariate classification with respect to ethnic background produced weak but statistically significant models. These results suggest that although NMR-based metabolomics can monitor changes in the urinary excretion profile of pregnant women, it may not be a prudent choice for the study of GDM.