ObjectiveTo build data analysis pipeline applied for the high-throughput sequencing data analysis process for the clinical diagnosis of genetic diseases. Methods87 WES data of multiple malformations cases from the biobank of neonatal birth defect of Children's Hospital of Fudan University (our hospital). Candidate mutation screening was performed using the pipeline of published whole-exome sequencing from Baylor in the USA (Baylor pipeline) as a reference standard. The same batch of data was then run using high-throughput sequencing data analysis and clinical diagnosis pipeline (Fudan pipeline). The results were then compared line by line using the chromosome location and the base position of the mutations as the identifiers. ResultsThe total number of the variants detected in 87 cases was 7 820 777. After the quality control, selective filtering from public database, in-house database, and screening from mutation type and Mendelian diseases associated variance filtering, an average of 134-210 variation was as the final candidate for clinical diagnosis analysis. 15 242 candidate variations were filtered from Baylor pipeline, and 15 660 from Fudan pipeline, both of them hit 15 137 variations. 523 more variations were found by Fudan pipeline, but there were still 105 variants missed, which hit in Baylor pipeline. When compared with Baylor pipeline, the sensitivity, specificity, positive predictive percentage and negative predictive percentage were all higher than 96.66%. ConclusionFudan pipeline could simulate Baylor pipeline for the performing of WES data filtering, effectively acquire candidate variation. Though the current Fudan pipeline is only a basic framework, it provides an indispensable process of using WES in clinical practice.
Objective To assess whether G71R mutation of the UGT1A1 gene is associated with severe neonatal hyperbilirubinemia of unknown cause.
Methods In a case-control study performed at a single hospital center in China, cases with severe hyperbilirubinemia （defined as bilirubin level≥342 μmol·L-1）and controls from neonatal birth defects samples database（bilirubin level <221 μmol·L-1) were enrolled. Polymerase chain reaction analysis of blood was performed to determine the frequency of UGT1A1 mutation in cases. All subjects were born with gestational age ≥35 weeks and birth weight ≥2500 g.
Results A total of 65 severe hyperbilirubinemic neonates and 65 control neonates were enrolled into the study. UGT1A1 G71R was the most common mutation in cases（48/65）. No statistical difference in the genotype or 211 G to A allele frequencies was found between controls and Chinese healthy infants from a meta-analysis(P＞0.05).The allelic frequency of "A" in G71R was 0.5 in severe hyperbilirubinemic neonates, which was significantly higher than 0.15 in the control group (P<0.001), power was 0.993.The genotype frequency of "A/A+G/A" was 73.8% in severe hyperbilirubinemic neonates, which was significantly higher than 27.7% in the control group (P<0.001), power was 1. The result revealed that the G71R mutant carriers (A/A+G/A) were associated with an increased risk of severe neonatal hyperbilirubinemia compared with the G/G allele carriers (OR=7.373, 95%CI:3.395-16.008)
Conclusion This finding strongly suggests that the presence of the G71R mutation contributes to the development of neonatal severe hyperbilirubinemia of unknown etiologyin the Chinese population.
ObjectiveIn this study, the whole-exome sequencing was used to identify pathogenic mutations in a newborn with frontal bossing, high-arched palate, low set ear, ventriculomegaly, absent septum pellucidum and agenesis of the corpus callosum. MethodsThe exome targets of the patient's DNA were captured with the SureSelct Human All Exon kit followed by sequencing with the Illumina HiSeq2000 platform. The data analysis pipeline of the Children's Hospital of Fudan University was used for the annotation and variant calling. The detected variant was confirmed with Sanger direct sequencing. ResultsWhole-exome sequencing identified total 79 064 variants in the proband. After filtering of quality, frequency criteria and classification, 645 rare variants were identified. Among these rare variants, 159 variants were mapped to disease-related genes in HGMD or OMIM database. Manual analysis revealed a reported heterozygous mutation (c.C1040G, p.S347C) in FGFR2 gene. Sanger direct sequencing confirmed that it was a de novo mutation in the proband. ConclusionWe present a patient with FGFR2 related disorder detected by whole-exome sequencing. This study clearly shows the efficacy of Whole-exome sequencing and our data analysis pipeline for rapid genetic diagnosis of multiple congenital malformations in newborn with an unknown cause.
ObjectiveTo evaluate the application value of targeted next generation sequencing for the molecular diagnosis in children with high risk of inherited metabolic disorders. MethodsFTA cards with dried blood spots were collected from the patients coming to genetic and metabolic specialist outpatient clinic, whose mass spectrometry analysis(MAS) was abnormal. Next generation sequencing (NGS) of a panel of 89 genes known to cause genetic disorders in children was uesed to find the genetic cause of the diseases. All mutations detected by NGS were subsequently confirmed by Sanger sequencing. ResultsA total of 48 patients with abnormal MAS findings were recruited retrospectively from June 2013 to May 2014. Forty patients completed the study finally, including 22 males. Among those samples, 29 samples accounting for 72.5 percent of the total samples, were detected mutations on the candidate genes that were consistent with the clinical symptoms, including 6 methylmalonic acidemia, 6 citrin deficiency, 5 phenylketonuria, 3 glutaric acidemia and 9 miscellaneous genetic disorders. Single pathogenic mutation was found in gene SLC25A13 by NGS in 2 suspected citrin deficiency cases. Long range PCR was done after that and another 3 kb insertion mutation was found in both of the cases. ConclusionApplication of targeted next generation sequencing in children with high risk of inherited metabolic disorders can provide reliable molecular diagnosis in a cost- and time-efficient manner. Identification of disease-causing mutations may have benefit in clinical practice and is essential for genetic counseling.
Objective Birth weight has effect not only on growth of infants but also on later life. Every year, children will be given routine examination by government at community health service centers in Minhang District in Shanghai, China. To observe the effect of birth weight on risk of overweight and obesity among children aged 5 to 8 years in China. Methods A retrospective longitudinal study was performed. Birth weight, current height and weight, and other anthropometric data of children were obtained from the community health service centers. Results Among 99 897 subjects, 10.1% of the children were overweighted and 5.0% were obese. Of all the children, 1.4% had low birth weight (birth weight <2 500 g) and 7.3% had high birth weight(birth weight≥4 000 g). The percentages of overweight were 8.4%, 7.6%, 9.2%, 11.5%, 14.5%, respectively, and the percentages of obesity were 4.7%, 3.9%, 4.3%, 5.9%, 7.9% respectively, across the birth weight categories:<2 500 g, -2 500 g, -3 000 g, -3 500 g, -4 000 g, ≥4 000 g. Comparing with the control group which was defined the birth weight as more than or equal to 3 000 g and less than 3 500 g, the risk ratio of the children with birth weight between 3 500 and 4 000 g for overweight increased 31% and for obesity increased 43%(RR=1.31, 95%CI:1.24-1.38, P<0.000 1; RR=1.43, 95%CI:1.32-1.54, P= 0.000 1, respectively) and the risk ratio of the children with birth weight equal to or more than 4 000 g for overweight increased 76% and for obesity increased 103%(RR=1.76, 95%CI:1.62-1.92, P<0.000 1;RR=2.03, 95%CI:1.81-2.23, P<0.000 1, respectively). The association was persisted after adjustment of the confounders including age, sex, birth height, pregnant week, delivery models, current height and breast feeding. The RRs of overweight and obesity were close among children aged 5, 6, 7, 8 years with birth weight of -4 000 g and ≥4 000 g. The children with birth weight more than or equal to 2 500 g and less than 3 000 g had no risk for overweight and obesity when they were 5 to 8 years old(RR=0.80, 95%CI, 0.73-0.88, P<0.000 1; RR=0.87, 95%CI:0.77-0.98, P<0.000 1), compared with the control group. The risk ratio for obesity increased to 1.07%-1.17% when birth weight decreased to less than 2 500 g, although the P values were not significant (crude RR=1.07, 95%CI:0.79-1.43, P= 0.27, adjusted RR=1.17, 95%CI:0.87-1.58, P= 0.68, respectively.). Conclusion A non-linear relationship between birth weight and risk of overweight and obesity was found in the children aged 5-8 years in Shanghai, China. The children with high birth weight were at an increased risk to have overweight and obesity at school-age. The children with low birth weight had the tendency of obesity when they were at 5 to 8 years age.
Objective To establish the reference intervals of lactate dehydrogenase(LDH), creatine kinase(CK), creatine kinase MB(CK-MB) and aspartate aminotransferase(AST) based on dry chemical system in healthy pediatric population aged from 3 months to 18 years in Shenzhen. Methods Healthy pediatric population aged from 3 months to 18 years were divided into four groups by age according to the rule of growth and development:3 months to 3 years,-6 years,-12 years and -18 years age groups.The serum samples were from the remaining specimens of routine physical examination after routine testing.Serum LDH,CK,CK-MB and AST were detected by the VITROS 350 system,and the differences between different age groups and different genders were compared and the reference intervals of 4 items among healthy children and adolescents aged from 3 months to 18 years were established. Results A total of 2 345 qualified specimens from children and adolescents were recruited, including 1 346 males and 999 females, with 187, 918, 916 and 324 subjects for 3 months to 3 years, -6 years, -12 years and -18 years age groups. ①The reference intervals of LDH showed no significant difference between boys and girls (P>0.05), but the CK of -6 years boys, CK, AST of -12 years boys, CK, AST, CK-MB of -18 years old boys were significantly higher than those in girls (P<0.05). ② The 4 items showed significant differences in different age groups (P<0.05), but the CK showed no significant difference between 3 months to 3 years group and -18 years group, the CK-MB showed no significant difference between 3 months to 3 years group and -6 years group, the AST showed no significant difference between -6 years group and -12 years group (P>0.05). The reference intervals of serum LDH (U·L-1): 3 months to 3 years 504-896, 3-12 years 491-790, 12-18 years 342-716; reference intervals of CK (U·L-1):3 months to 12 years 50-226, 12-18 years 47-256 (male), 34-235 (female); reference intervals of CK-MB (U·L-1): 3 months to 6 years 18-44, 6-12 years 7-35, 12-18 years 7-35; reference intervals of AST (U·L-1): 3 months to 3 years 23-60, 3-12 years 24-27, 12-18 years 17-47 U·L-1. ③The reference intervals of 4 items overall decrease with age. Conclusion Dry chemical system of minor myocardial enzymes reference intervals are unique and significantly associated with the change of age.
Objective A systematic review was conducted to assess the influence of topiramate on children′s cognitive function to provide high quality evidence for clinical practice. Methods The literatures were searched using PubMed, Cochrane Library, EMBASE(Ovid), China Biology Medicine disc(CBM), China National Knowledge Infrastructure (CNKI), WanFang Data, VIP database from establishment until now. Two researchers independently screened the studies using predetermined inclusion and exclusion criteria and assessed the quality of these studies and extracted the data. Quantitative analysis of cognitive-related ADEs was performed using meta-analysis with non-comparative binary data, the outcome measured by different scales was represented. Results 676 studies were found and 9 studies were finally included in this systematic review, 3 RCTs and 6 case series, including 1 240 patients. Among 9 experimental studies, 2 RCTs and 6 case series indicated low risk bias, 1 RCT had high risk bias .The incidence and 95% confidence interval of cognitive-related ADEs: confusion 0.03(0.01-0.06) , psychomotor slowing 0.08(0.02-0.26), difficulty with concentration/attention 0.07(0.02-0.21), word-finding difficulties0.07(0.04-0.14), depression or mood problems 0.05(0.02-0.11), fatigue 0.05(0.02-0.10),somnolence 0.12(0.04-0.28), difficulty with memory and speech or language problems were without statistical significance. Cognitive function was worsen treated by TPM than VPA,topiramate 100 mg per day was associated with increases in psychomotor reaction times but memory was unchanged, there were no significant differences between TPM monotherapy and converting monotherapy. Conclusion The incidence of cognitive-related ADEs was low in children received topiramate treatment. Topiramate was associated with modest increases in psychomotor reaction times. Learning and memory were unchanged in migraine children.
Objective To detect some virulence genes of the Escherichia coli(E. coli) isolated from hospitalized neonates. Methods Subjects were recruited based on consecutive cases who were diagnosed as pneumonia and septicemia from neonates hospitalized in the Neonatal Center of Beijing Children′s Hospital from September 2009 to May 2012. Before using antibiotics sputum, blood, cerebrospinal fluid, or navel secretion cultivation were selected for cases of E. coli. According to the onset age, strains were divided into early-onset or late onset. According to different sources of specimens, strains were divided into invasive infection and non invasive infection. Polymerase chain reaction (PCR) method was used to detect the 11 most common virulence genes of Escherichia coli, including heat-labile (LT1) ,heat-stable (ST1 and ST2) toxins, verotoxin types1, 2, 2e (VT1, VT2 and VT2e, respectively), cytotoxic necrotizing factors (CNF1 and CNF2), attaching and effacing mechanisms (eaeA), enteroaggregative mechanisms (Eagg),and enteroinvasive mechanisms (Einv). Chromas software was used to read sequence, the sequencing results directly performed for BLAST searching by using Chromas software. Results A total of 110 Escherichia coli strains were isolated from extraintestinal sites in 102 neonates. From respiratory secretions (sputum) 79 strains (71.8%) were isolated, and blood 18 strains (16.4%), cerebrospinal fluid 6 strains (5.4%), umbilical discharge 7 strains (6.4%). Among the involved neonates, 102 were from full-term infants, 8 strains from preterm neonates. There were 27 early-onset strains, 83 late onset strains. 24 strains of invasive infection, 86 strains of non invasive infection. Sixty-seven cases were diagnosed as pneumonia and 35 cases as neonatal septicemia. ① One hundred and one strains (91.8%) were positive for Einv gene, 24 strains (21.8%) were positive for CNF1 gene, 14 strains (12.7%) were positive for CNF2 gene, 1 strain (0.9%) was positive for Eagg gene. ②Among 110 strains, there were 7 different toxin gene profiles. Only carrying Einv gene was the most common, harbored by 66 isolates (60%).One strain (0.9%) carried CNF1 gene, 20 strains (18.2%) carried Einv and CNF1 genes, 11 strains (10.0%) carried Einv and CNF2 genes, 3 strains (2.7%) carried Einv, CNF1 and CNF2 genes, only 1 strain (0.9%) carried Einv and Eagg gene, 8 strains of them had none of the 11 virulences（7.3%）. ③The virulence gene carrying rate and different toxin gene profiles of E. coli strains did not significantly differ in sites or disease onset age. There were 37 cases carring CNF1 or CNF2 gene, of them, 22 cases with respiratory failure, 15 cases needed NCPAP respiratory support, 10 cases with multiple organ damage, 3 cases with pulmonary hemorrhage. ④There were 8 pairs of clinical isolates from same neonatal patient, but they were isolated from different time or different sites. Seven pairs of clinical isolates belonged to the same genotype after PFGE typing, and 3 pairs of strain were isolated from different times of the same neonate, PFGE typing result was consistent, but the virulence gene carrying mode was not the same. Conclusion This study showed that PCR method could be used to detect 11 virulence genes, but only 4 genes were detected. The most abundant toxin gene was Einv. There were 7 different toxin gene profiles. Only carrying Einv gene was the most common. CNF1 and CNF2 gene may be related with severity of pneumonia.
Objective To realize the efficacy and safety of pegylated interferon-alpha or interferon-alpha plus ribavirin in the treatment of children with hepatitis C genotype 1b. Methods Clinical manifestations, baseline characteristics, hepatitis C virus RNA levels during treatment and adverse events were retrospectively analyzed children with hepatitis C genotype 1b who visited Children's Hospital of Fudan University from November 2011 to December 2014. Results Ten children patients with hepatitis C genotype 1b were recruited, including 8 boys and 2 girls. The mean age of the patients was 37.1 months (8 months to 6.6 years old). All children had no obvious clinical manifestations, 2 were infected by mother-to-child and 8 were infected by other uncertain ways. ①Before treatment, 9 patients with hepatitis C antibody positive, HCV viral loads were higher than 1×103IU·mL-1 in 8 patients; 5 children had mild alanine aminotransferase and aspartate transaminase elevated. Liver biopsies were performed in 7 children, pathological results showed mild steatosis, liver tissue inflammation grade G1 to G3, liver tissue fibrosis stage S0 to S2. ②Four children were treated with pegylated interferon-alpha plus ribavirin and three children were treated with interferon-alpha plus ribavirin. Six patients became HCV RNA negative (<1×103 IU·mL-1) after 12-week treatment (early virological response, EVR) and kept HCV RNA negative during 36-48 treatment weeks. One patient became HCV RNA negative after 4-week treatment, and was still in treatment. Two cases who maintained HCV RNA negative and normal liver function received no antiviral treatment. ③No severe adverse events were observed in the 7 children who received treatment. Pyrexia occurred in all of the children at the beginning of treatment. After 4-week treatment, 1 child had a moderate reduce of white blood cells, platelets decreased transiently three times and hemoglobin level remained normal. Conclusion Liver tissue inflammation and fibrosis are relatively mild in the children with hepatitis C genotype 1b. The regimen consisted of pegylated interferon-alpha or interferon-alpha and ribavirin achieves a good response in our children with hepatitis C genotype 1b. No severe adverse events occurred and most of the patients well tolerate the therapy.
Objective To observe the CD2AP expression in renal tissue of children with primary nephrotic syndrome(PNS), and to discuss the association between the expression of CD2AP and podocytes injury． Methods Children with PNS after renal biopsy in Shanghai Children′s Hospital from Jan. 2012 to Jun. 2013 were enrolled as PNS group, patients with renal tumor undergone nephrectomy were recruited as control group. Renal pathological change was observed by conventional optical microscopes. Structure of podocyte was observed by electron microscopy. The renal CD2AP mRNA and protein were examined by quantitive real time PCR and immunohistochemistry. Then serum albumin（Alb）,creatinine（Cr）,complement 3（C3）,triglyeride（TG）,cholesterol（TC）,glomerular filtration rate（GFR）and 24-hours proteinuria were detected, simultaneously the relationship between these index and the CD2AP expression was analyzed. Results Eighteen patients were enrolled as PNS group and 11 patients were enrolled as control group. ①Compared to the controls, the expression of CD2AP mRNA decreased significantly in patients with PNS（0.008 58±0.006 12 vs 0.019±0.012, P＜0.05）. ②Compared to the control group, protein expression decreased significantly in PNS group, (1.302±0.885) vs (2.24±1.18), P＜0.05, and children with PNS were the lowest. ③In PNS group, there was moderate correlation between CD2AP mRNA relative expression and TG（r=-0.527，P＜0.05）, 24 h proteinuria（r=-0.602，P＜0.01）, there was strong positive correlation between CD2AP mRNA and protein expressions (r=0.788, P＜0.01). The relationship between CD2AP relative expression and those index such as complement C3, GFR, Cr and TC had no statistic significance. Conclusion The expression of CD2AP in podocyte of PNS cases decreased significantly, especially in those with FSGS. And this change may be correlated with clinical indexes of patients with PNS.
