基于Web Service的生物学实体映射数据库

生物学实体映射就是要实现对基因、蛋白质、小分子物质、化合物和药物等实体的不同标识符和名称之间的相互转换.生物学实体映射可以帮助生物医学研究者将实验结果关联到海量的在线生物医学数据资源,并为生物医学文本挖掘和信息检索的研究者在命名实体识别和查询关键词扩展方面提供技术支持.构建一个生物学实体映射数据库,其中存储了大量的生物学实体映射信息;还构建一个基于Web Service的生物学实体映射网络应用系统,为用户同时提供通过浏览器和通过Web Service的两种方式访问生物学实体映射数据库.     

基因组扫描技术在医学遗传学研究中的应用

随着人类基因组计划的进行 ,各种疾病基因的定位、克隆已成为今后医学遗传学研究的重点。而基因组扫描技术正是致力于寻找疾病相关新基因的一种有效手段。 2 0世纪 90年代以来 ,随着微卫星 DNA技术成功用于基因组扫描研究 ,使基因组扫描技术由最初的手工操作发展到目前的自动化操作。基因组扫描技术的发展 ,不仅大大加快了人类单基因遗传病相关基因定位的速度 ,而且也使人类的高血压、糖尿病、肿瘤、精神分裂症等复杂性状疾病相关基因的定位和分离取得重大突破     

NetRD:一种利用Bing搜索结果补充文献挖掘证据集的工具

目的当前生物文献挖掘工作的重心是改进各挖掘模块性能,以提升挖掘结果的可信度,但有很大比例的挖掘结果其文献证据很少,为此本文提出一个利用Bing搜索引擎从海量Web数据中为文献挖掘得到的生物实体关联对提供补充证据的工具系统。方法利用现有文本挖掘技术从PubMed文献中挖掘一批生物实体关联对,引入BingWeb搜索模块,以生物实体名作为关键词从Web中利用Bing开放搜索API得到一批搜索结果,将这些结果整理成新的数据源,最终从该新的数据源中挖掘得到一批来自Web的补充证据。结果本系统(http://bioinfo.ustc.edu.cn/NetRD)对文献证据较少的生物实体关联对提供了有效的补充证据支持,丰富了文献挖掘结果最终的证据集。结论以Web数据作为补充数据源,能够有效地为文献证据很少的生物实体对提供证据补充,为相关研究者确认两个生物实体之间的关联提供重要参考。     

基因组扫描技术在医学遗传学研究中的应用

随着人类基因组计划的进行，各种疾病基因的定位，克隆已成为今后医学遗传学研究的重点，而基因组扫描技术正致力于寻找疾病相关新基因的一种有效手段，20世纪90年代以来，随着微卫生DNA技术成功用于基因组扫描研究，使基因组扫描技术由初的手工操作发展到目前的自动化操作，基因组扫描技术的发展，不仅大大加快了人类单基因遗传病相关基因定位的速度，而且也使人类的高血压，糖尿病，肿瘤，精神分裂症等复杂性疾病相关基因的定位和分离取得重大突破。     

基于MR T2加权成像的影像组学标签预测直肠癌KRAS基因突变的价值

目的 探讨基于MR T₂加权成像(T₂WI)的影像组学标签预测直肠癌KRAS基因突变的潜在价值。方法 回顾性研究。纳入山西省肿瘤医院2017年4月—2019年4月行盆腔MR检查并具有KRAS基因检测结果的304例直肠癌患者的临床和影像资料,其中男175例、女129例,中位年龄59.6岁。按7∶3比例将患者随机分为训练组(213例)和验证组(91例)。选取每例患者的高分辨率T₂WI进行图像分割及影像组学特征提取,使用单变量统计分析为主的“五步法”进行特征降维,并分别采用多变量logistic回归、决策树(DT)以及支持向量机(SVM)三种分类算法构建影像组学标签,用于预测直肠癌KRAS基因状态。受试者操作特征(ROC)曲线、校正曲线、决策曲线分析(DCA)评估影像组学标签的预测性能及临床效益。结果 训练组和验证组患者的基线资料比较以及两组中KRAS突变型与野生型患者的临床特征比较,差异均无统计学意义(P值均＞0.05)。从每位患者的T₂WI中提取960个影像组学特征,经特征筛选后得到7个与直肠癌KRAS基因相关的特征(P值均＜0.05)。采用多变量logistic回归、DT及SVM构建的三个预测模型的ROC曲线下面积,训练组分别为0.677、0.604和0.722,验证组分别为0.626、0.600和0.682,其中SVM模型在预测KRAS基因状态方面效能最好。DCA曲线示三种预测模型均有一定的临床效益,其中SVM预测模型净收益值最大。结论 基于MR T₂WI的影像组学标签在预测直肠癌KRAS基因状态方面有一定的价值。     

糖尿病遗传标志的分子生物学研究进展

分子生物学技术为糖尿病遗传学研究开辟了广阔前景。应用此项技术研究糖尿病遗传标志见到:胰岛素依赖型糖尿病(IDDM)与人白细胞抗原D(HLAD)位点的DRα、DRβ、DQβ、DXα基因关联,尚提示与T细胞受体β基因、免疫球蛋白重链基因及胰岛素基因关联;非胰岛素依赖型糖尿病(NIDDM)与胰岛素受体基因、载脂蛋白AI基因及载脂蛋白B基因关联。识别糖尿病遗传标志,从而阐明糖尿病分子遗传学发病机制,将有助于在群体中早期检出高危人群。此在糖尿病防治中有重要意义。     

一种与糖尿病有关的新基因被发现

《Nature》报道,一种新近与1型糖尿病的发病关联起来的基因的发现,可能会推动糖尿病预测试验的发展。对来自超过500名1型糖尿病患者和由超过1000名志愿者组成的对比组的DNA所进行的一项整个基因组范围内的关联研究,证实了患者存在一种新的基因,其与已知的1型糖尿病相关基因和K     

基于背景估计和SVM分类器的眼底图像硬性渗出物检测方法

硬性渗出物是糖尿病视网膜病变(DR)的早期病症,是糖尿病性黄斑水肿的最主要表现,因此对硬性渗出物的准确检测具有重要的临床意义。提出一种基于背景估计和SVM分类器的眼底图像硬性渗出物检测方法。首先通过背景估计,得到包含亮目标的前景图;然后利用基于Kirsch算子的边缘信息确定硬性渗出物的候选区域,再移除视盘;最后对候选区域进行形状特征、直方图统计特征以及相位特征的提取,采用SVM对候选区域进行分类,完成硬性渗出物的精确提取。对DIARETDB1和HEI MED公共数据库中共248幅眼底图像进行实验,图像水平达到灵敏度97.3%和特异性90%,病灶水平达到灵敏度84.6%和阳性预测值944%。实验表明,所提出的方法能够实现眼底图像中硬性渗出物的自动检测。     

基于文本挖掘的流行病学致病因素的提取

目的 基于文本挖掘技术,设计出能够自动提取流行病学致病因素的系统.方法 该自动信息提取系统由一个文本挖掘引擎子系统和一个基于规则的信息提取子系统构成.首先使用文本挖掘引擎标记出所有的名词短语,并收集该名词短语的语义等信息.然后利用基于规则的文本分类器,标记出流行病学致病因素.结果 为评估本系统,将由流行病学专家人工注解的文本输入该系统,评估发现最好的结果F-measure为64.6％,其精确率和召回率分别为61.0％和68.8％,该结果优于其它相关研究,且其中有些错误仍可避免.结论 基于文本挖掘的方法对从流行病学研究文献中自动提取致病因素信息有很大帮助.     

基于监督局部线性嵌入方法的阿尔茨海默病磁共振成像分类研究

针对阿尔茨海默病(AD)早期阶段分类这一研究难题,传统的线性特征提取算法很难从其高维特征中挖掘出鉴别能力较强的信息来有效地表示样本特征。因此,本文采用监督局部线性嵌入(SLLE)特征提取算法,对412例受试者的大脑皮质厚度(CTH)和脑感兴趣区域体积(VOI)特征进行提取,减少其冗余特征以提高识别精度。受试者来源于阿尔茨海默病神经影像学(ADNI)数据集,包含93例稳定型轻度认知障碍(s MCI)、96例遗忘型轻度认知障碍(a MCI)、86例AD患者和137例认知正常对照老年人(CN)样本。本文采用的SLLE算法是通过添加距离修正项来计算每个样本点的近邻点,并用近邻点线性表示样本,得到局部重建权值矩阵,进而求出高维数据的低维映射。为验证该算法在分类识别中的有效性,本文将主成分分析(PCA)、近邻最小最大投影(NMMP)、局部线性映射(LLE)及SLLE等特征提取算法分别与支持向量机(SVM)分类器组合,对CN与s MCI、CN与a MCI、CN与AD、s MCI与a MCI、s MCI与AD和a MCI与AD六组实验数据进行分类识别。结果显示,以VOI为特征,利用SLLE和SVM的复合算法对s MCI和a MCI的分类准确度、灵敏度、特异性分别为65.16%、63.33%、67.62%,基于LLE和SVM的复合算法分类结果分别为64.08%、66.14%、62.77%,而基于传统SVM则分别为57.25%、56.28%、58.08%。经比较,发现SLLE和SVM组合算法的识别精度较LLE和SVM的组合算法提高了1.08%,较SVM提高了7.91%。因此,利用SLLE和SVM这一复合算法进行分类识别更有利于AD的早期诊断。     

全基因组筛查与2型糖尿病相关基因

2型糖尿病是一种多基因遗传性疾病，其发病是不同基因之间相互作用，基因与环境之间相互作用的结果。利用全基因组筛查与连锁分析的方法，目前已在多条染色体上确定了比如1q21-q24、2q37、3q27、4q、5q34-q35.2、6q21-23、12q24、20q12-13.1等多个染色体区域与2型糖尿病相关，并且在这些区域内发现了INSRR、PKLR 、CAPN10、HNF1A、GLUT10等作为2型糖尿病易感基因的候选基因。     

HLA DRB1, DQB1 and insulin promoter VNTR polymorphisms: interactions and the association with adult-onset diabetes mellitus in Czech patients

Both the human leucocyte antigen (HLA) DRB1 and the HLA DQB1 gene loci play a role in the development and progression of autoimmune diabetes mellitus (T1DM). Similarly, the insulin promoter variable number tandem repeats (INS-VNTR) polymorphism is also involved in the pathogenesis of diabetes mellitus (DM). We studied the association between each of these polymorphisms and DM diagnosed in patients older than age 35 years. Furthermore, we analysed possible interactions between HLA DRB1/DQB1 and INS-VNTR polymorphisms. Based on C-peptide and GADA levels we were able to distinguish three types of diabetes: T1DM, latent autoimmune diabetes in adults (LADA) and T2DM. INS-VNTR was genotyped indirectly by typing INS-23HphI A/T polymorphism. The genotype and allele frequencies of INS-23HphI did not differ between each of the diabetic groups and group of healthy subjects. We did, however, observe an association between the INS-23HphI alleles, genotypes and C-peptide secretion in all diabetic patients: A allele frequency was 86.2% in the C-peptide-negative group vs. 65.4% in the C-peptide-positive group (P(corr.) < 0.005); AA genotype was found to be 72.4% in the C-peptide-negative group vs. 42.6% in the C-peptide-positive groups (P(corr.) < 0.01). The HLA genotyping revealed a significantly higher frequency of HLA DRB1*03 allele in both T1DM and LADA groups when compared to healthy subjects: T1DM (25.7%) vs. control group (10.15%), odds ratio (OR) = 3.06, P < 0.05; LADA (27.6%) vs. control (10.15%), OR = 3.37, P < 0.01. The simultaneous presence of both HLA DRB1*04 and INS-23HphI AA genotype was detected in 37.5% of the T1DM group compared to only 9.2% of the healthy individuals group (OR = 5.9, P(corr.) < 0.007). We summarize that in the Central Bohemian population of the Czech Republic, the INS-23HphI A allele appears to be associated with a decrease in pancreatic beta cell secretory activity. HLA genotyping points to at least a partial difference in mechanism, which leads to T1DM and LADA development as well as a more diverse genetic predisposition in juvenile- and adult-onset diabetes. The simultaneous effect of HLA and INS-VNTR alleles/genotypes predispose individuals to an increased risk of diabetes development.     

A shallow parser based on closed-class words to capture relations in biomedical text

Natural language processing for biomedical text currently focuses mostly on entity and relation extraction. These entities and relations are usually pre-specified entities, e.g., proteins, and pre-specified relations, e.g., inhibit relations. A shallow parser that captures the relations between noun phrases automatically from free text has been developed and evaluated. It uses heuristics and a noun phraser to capture entities of interest in the text. Cascaded finite state automata structure the relations between individual entities. The automata are based on closed-class English words and model generic relations not limited to specific words. The parser also recognizes coordinating conjunctions and captures negation in text, a feature usually ignored by others. Three cancer researchers evaluated 330 relations extracted from 26 abstracts of interest to them. There were 296 relations correctly extracted from the abstracts resulting in 90% precision of the relations and an average of 11 correct relations per abstract.     

基于时域太赫兹波谱法的降糖类药物分类与分析

目的根据降糖类药物在太赫兹频域范围的吸收光谱,研究他们各自基团的振动频率及其共振峰,以有效鉴别各种相似药品,同时也为明确不同分子基团振荡在药理上的贡献打下基础。方法利用时域太赫兹波谱系统对格列喹酮、格列吡嗪、格列齐特、格列美脲、瑞格列奈和二甲双胍6种降糖药进行检测,得到他们在0.3～3.0 THz范围内的吸收谱;同时选取1.5～2.0 THz范围内的分类特征数据,用支持向量机的方法,分别对这些降糖药的特征数据进行鉴别。结果 4种磺脲类降糖药都在1.37 THz处有一明显的共振峰。根据吸收谱,能够很轻易地区分出瑞格列奈和二甲双胍与磺脲类药物的差别;而在支持向量机的帮助下,4类磺脲类药物也可被100%准确区分出来。结论太赫兹作为一种新的检测手段,在药物鉴别、药物品质控制、化学键/功能基团识别等方面有着积极的意义。     

Using semantic predications to uncover drug–drug interactions in clinical data

In this study we report on potential drug–drug interactions between drugs occurring in patient clinical data. Results are based on relationships in SemMedDB, a database of structured knowledge extracted from all MEDLINE citations (titles and abstracts) using SemRep. The core of our methodology is to construct two potential drug–drug interaction schemas, based on relationships extracted from SemMedDB. In the first schema, Drug1 and Drug2 interact through Drug1’s effect on some gene, which in turn affects Drug2. In the second, Drug1 affects Gene1, while Drug2 affects Gene2. Gene1 and Gene2, together, then have an effect on some biological function. After checking each drug pair from the medication lists of each of 22 patients, we found 19 known and 62 unknown drug–drug interactions using both schemas. For example, our results suggest that the interaction of Lisinopril, an ACE inhibitor commonly prescribed for hypertension, and the antidepressant sertraline can potentially increase the likelihood and possibly the severity of psoriasis. We also assessed the relationships extracted by SemRep from a linguistic perspective and found that the precision of SemRep was 0.58 for 300 randomly selected sentences from MEDLINE. Our study demonstrates that the use of structured knowledge in the form of relationships from the biomedical literature can support the discovery of potential drug–drug interactions occurring in patient clinical data. Moreover, SemMedDB provides a good knowledge resource for expanding the range of drugs, genes, and biological functions considered as elements in various drug–drug interaction pathways.     

染色体1q21-q25基因变异与2型糖尿病关系研究进展

2型糖尿病是一种多因素复杂疾病,遗传因素在其中占有很重要的地位.人类染色体1q21-q25区域富含与代谢、炎性反应、信号转导与基因转录等相关的基因,其与2型糖尿病的连锁关系已在不同人群中得到印证.本文就染色体1q21-q25区域的基因变异与2型糖尿病关系的研究进展进行综述.     

Genome-Wide Scan for CAG/CTG Repeat Expansions in Pimas with Early Onset of Type 2 Diabetes Mellitus

The expansion of polymorphic CAG/CTG repeats in specific genes causes several neurodegenerative disorders and in many instances the length of the disease-causing repeat correlates with the onset age and/or severity of symptoms. Type 2 diabetes mellitus has features in common with diseases resulting from trinucleotide repeat expansion, including a variable age of disease onset and penetrance. We have investigated whether CAG/CTG repeat expansion contributes to the genetic etiology of type 2 diabetes in the Pima Indians, a population with the highest reported prevalence of this disease. Using the Repeat Expansion Detection (RED) method, we determined the size range in nondiabetic Pimas to be between (CAG)₂₀and (CAG)₁₃₀(mean repeat length = 195 bp), which is significantly larger than the mean size reported in Caucasians (150 bp). We compared the distribution of CAG/CTG repeat lengths among 40 Pimas with an early onset of type 2 diabetes (<22 years) and 38 nondiabetic subjects (>55 years). A 240-bp CAG/CTG RED product was found more frequently in early onset diabetics relative to nondiabetic controls (26% vs 11%), whereas a 210-bp band was more prominent in unaffected subjects (29% vs 13%); however, these differences were not statistically significant. In one Pima kindred, we also identified large RED products (≥360 bp) that displayed intergenerational instability among family members. However, these expansions were not associated with diabetes or any other clinical abnormalities in the carriers. We conclude that this unstable CAG/CTG repeat may represent a novel locus, consisting of large, but apparently nonpathogenic, unstable sequences.     

New recessive syndrome of microcephaly, cerebellar hypoplasia, and congenital heart conduction defect

We identified a two-branch consanguineous family in which four affected members (three females and one male) presented with constitutive growth delay, severe psychomotor retardation, microcephaly, cerebellar hypoplasia, and second-degree heart block. They also shared distinct facial features and similar appearance of their hands and feet. Childhood-onset insulin-dependent diabetes mellitus developed in one affected child around the age of 9 years. Molecular analysis excluded mutations in potentially related genes such as PTF1A, EIF2AK3, EOMES, and WDR62. This condition appears to be unique of other known conditions, suggesting a unique clinical entity of autosomal recessive mode of inheritance.     

Unsupervised biomedical named entity recognition: Experiments with clinical and biological texts

Named entity recognition is a crucial component of biomedical natural language processing, enabling information extraction and ultimately reasoning over and knowledge discovery from text. Much progress has been made in the design of rule-based and supervised tools, but they are often genre and task dependent. As such, adapting them to different genres of text or identifying new types of entities requires major effort in re-annotation or rule development. In this paper, we propose an unsupervised approach to extracting named entities from biomedical text. We describe a stepwise solution to tackle the challenges of entity boundary detection and entity type classification without relying on any handcrafted rules, heuristics, or annotated data. A noun phrase chunker followed by a filter based on inverse document frequency extracts candidate entities from free text. Classification of candidate entities into categories of interest is carried out by leveraging principles from distributional semantics. Experiments show that our system, especially the entity classification step, yields competitive results on two popular biomedical datasets of clinical notes and biological literature, and outperforms a baseline dictionary match approach. Detailed error analysis provides a road map for future work.     

Genetic heterogeneity between type 1 a and type lb insulin-dependent diabetes mellitus: HLA Class II and TAP gene analysis

The aim of this study was to compare the genetic susceptibility linked to the HLA Class II region genes of the Major Histocompatibility Complex in isolated insulin-dependent diabetes mellitus (la-IDDM) and insulin-dependent diabetes mellitus associated with another autoimmune endocrinopathy (lb-IDDM). HLA genes DRB1, DQA1 and DQB1 were studied at the genomic level, as well as genes TAP1 and TAP2. One hundred and seventy-nine la-IDDM diabetic patients were compared with 83 lb-IDDM patients. While it appeared that common genetic traits characterize diabetes regardless of the subtype (la or lb), certain features differentiate the two forms of IDDM. Extending the analysis of risk haplotypes DRB1*03 and DRB1*04 to TAP genes elicited a difference between la-IDDM and lb-IDDM patients. Haplo-type DRB1*03 was thus characterized in la-IDDM patients by a lower frequency of alleles TAP1-B (13.5%) and TAP2-B (16.2%), not found in lb-IDDM patients (33.3% for each allele). Likewise, haplotype DRB1*04 is characterized in lb-IDDM patients by a lower frequency of alleles TAP1-C (4.0%) and TAP2-B (8.0%) than in la-IDDM patients (22.2% and 25.9%, respectively). In total, this study showed that extending the characterization of HLA Class II haplotypes to TAP genes discriminates between the forms of diabetes restricted to a specific pancreatic affection and those reflecting a wider autoimmune disorder affecting several organs.