Chun-Nan Hsu

Publications of Chun-Nan Hsu

• Automatic morphological subtyping reveals new roles of caspases in mitochondrial dynamics.

  Authors: Jyh-Ying Peng, Chung-Chih Lin, Yen-Jen Chen, Lung-Sen Kao, Young-Chau Liu, Chung-Chien Chou, Yi-Hung Huang, Fang-Rong Chang, Yang-Chang Wu, Yuh-Show Tsai, Chun-Nan Hsu

  PLoS computational biology. 10/2011; 7(10):e1002212.

  Morphological dynamics of mitochondria is associated with key cellular processes related to aging and neuronal degenerative diseases, but the lack of standard quantification of mitochondrialMorphological dynamics of mitochondria is associated with key cellular processes related to aging and neuronal degenerative diseases, but the lack of standard quantification of mitochondrial morphology impedes systematic investigation. This paper presents an automated system for the quantification and classification of mitochondrial morphology. We discovered six morphological subtypes of mitochondria for objective quantification of mitochondrial morphology. These six subtypes are small globules, swollen globules, straight tubules, twisted tubules, branched tubules and loops. The subtyping was derived by applying consensus clustering to a huge collection of more than 200 thousand mitochondrial images extracted from 1422 micrographs of Chinese hamster ovary (CHO) cells treated with different drugs, and was validated by evidence of functional similarity reported in the literature. Quantitative statistics of subtype compositions in cells is useful for correlating drug response and mitochondrial dynamics. Combining the quantitative results with our biochemical studies about the effects of squamocin on CHO cells reveals new roles of Caspases in the regulatory mechanisms of mitochondrial dynamics. This system is not only of value to the mitochondrial field, but also applicable to the investigation of other subcellular organelle morphology.

• Interrogation of rabbit miRNAs and their isomiRs.

  Authors: Sung-Chou Li, Yu-Lun Liao, Wen-Ching Chan, Meng-Ru Ho, Kuo-Wang Tsai, Ling-Yueh Hu, Chun-Hung Lai, Chun-Nan Hsu, Wen-chang Lin

  Genomics. 09/2011; 98(6):453-9.

  Rabbit (Oryctolagus cuniculus) is the only lagomorph animal of which the genome has been sequenced. Establishing a rabbit miRNA resource will benefit subsequent functional genomic studies in mammals.Rabbit (Oryctolagus cuniculus) is the only lagomorph animal of which the genome has been sequenced. Establishing a rabbit miRNA resource will benefit subsequent functional genomic studies in mammals. We have generated small RNA sequence reads with SOLiD and Solexa platforms to identify rabbit miRNAs, where we identified 464 pre-miRNAs and 886 mature miRNAs. The brain and heart miRNA libraries were used for further in-depth analysis of isomiR distributions. There are several intriguing findings. First, several rabbit pre-miRNAs form highly conserved clusters. Second, there is a preference in selecting one strand as mature miRNA, resulting in an arm selection preference. Third, we analyzed the isomiR expression and validated the expression of isomiR types in different rabbit tissues. Moreover, we further performed additional small RNA libraries and defined miRNAs differentially expressed between brain and heart. We conclude also that isomiR distribution profiles could vary between brain and heart tissues.

• Soft tagging of overlapping high confidence gene mention variants for cross-species full-text gene normalization.

  Authors: Cheng-Ju Kuo, Maurice H T Ling, Chun-Nan Hsu

  BMC bioinformatics. 01/2011; 12 Suppl 8:S6.

  Previously, gene normalization (GN) systems are mostly focused on disambiguation using contextual information. An effective gene mention tagger is deemed unnecessary because the subsequent steps willPreviously, gene normalization (GN) systems are mostly focused on disambiguation using contextual information. An effective gene mention tagger is deemed unnecessary because the subsequent steps will filter out false positives and high recall is sufficient. However, unlike similar tasks in the past BioCreative challenges, the BioCreative III GN task is particularly challenging because it is not species-specific. Required to process full-length articles, an ineffective gene mention tagger may produce a huge number of ambiguous false positives that overwhelm subsequent filtering steps while still missing many true positives. We present our GN system participated in the BioCreative III GN task. Our system applies a typical 2-stage approach to GN but features a soft tagging gene mention tagger that generates a set of overlapping gene mention variants with a nearly perfect recall. The overlapping gene mention variants increase the chance of precise match in the dictionary and alleviate the need of disambiguation. Our GN system achieved a precision of 0.9 (F-score 0.63) on the BioCreative III GN test corpus with the silver annotation of 507 articles. Its TAP-k scores are competitive to the best results among all participants. We show that despite the lack of clever disambiguation in our gene normalization system, effective soft tagging of gene mention variants can indeed contribute to performance in cross-species and full-text gene normalization.

• The gene normalization task in BioCreative III.

  Authors: Zhiyong Lu, Hung-Yu Kao, Chih-Hsuan Wei, Minlie Huang, Jingchen Liu, Cheng-Ju Kuo, Chun-Nan Hsu, Richard Tzong-Han Tsai, Hong-Jie Dai, Naoaki Okazaki [......] Fabio Rinaldi, Sanmitra Bhattacharya, Padmini Srinivasan, Hongfang Liu, Manabu Torii, Sergio Matos, David Campos, Karin Verspoor, Kevin M Livingston, W John Wilbur

  BMC bioinformatics. 01/2011; 12 Suppl 8:S2.

  We report the Gene Normalization (GN) challenge in BioCreative III where participating teams were asked to return a ranked list of identifiers of the genes detected in full-text articles. ForWe report the Gene Normalization (GN) challenge in BioCreative III where participating teams were asked to return a ranked list of identifiers of the genes detected in full-text articles. For training, 32 fully and 500 partially annotated articles were prepared. A total of 507 articles were selected as the test set. Due to the high annotation cost, it was not feasible to obtain gold-standard human annotations for all test articles. Instead, we developed an Expectation Maximization (EM) algorithm approach for choosing a small number of test articles for manual annotation that were most capable of differentiating team performance. Moreover, the same algorithm was subsequently used for inferring ground truth based solely on team submissions. We report team performance on both gold standard and inferred ground truth using a newly proposed metric called Threshold Average Precision (TAP-k). We received a total of 37 runs from 14 different teams for the task. When evaluated using the gold-standard annotations of the 50 articles, the highest TAP-k scores were 0.3297 (k=5), 0.3538 (k=10), and 0.3535 (k=20), respectively. Higher TAP-k scores of 0.4916 (k=5, 10, 20) were observed when evaluated using the inferred ground truth over the full test set. When combining team results using machine learning, the best composite system achieved TAP-k scores of 0.3707 (k=5), 0.4311 (k=10), and 0.4477 (k=20) on the gold standard, representing improvements of 12.4%, 21.8%, and 26.6% over the best team results, respectively. By using full text and being species non-specific, the GN task in BioCreative III has moved closer to a real literature curation task than similar tasks in the past and presents additional challenges for the text mining community, as revealed in the overall team results. By evaluating teams using the gold standard, we show that the EM algorithm allows team submissions to be differentiated while keeping the manual annotation effort feasible. Using the inferred ground truth we show measures of comparative performance between teams. Finally, by comparing team rankings on gold standard vs. inferred ground truth, we further demonstrate that the inferred ground truth is as effective as the gold standard for detecting good team performance.

• UMARS: Un-MAppable Reads Solution.

  Authors: Sung-Chou Li, Wen-Ching Chan, Chun-Hung Lai, Kuo-Wang Tsai, Chun-Nan Hsu, Yuh-Shan Jou, Hua-Chien Chen, Chun-Hong Chen, Wen-Chang Lin

  BMC bioinformatics. 01/2011; 12 Suppl 1:S9.

  Un-MAppable Reads Solution (UMARS) is a user-friendly web service focusing on retrieving valuable information from sequence reads that cannot be mapped back to reference genomes. Recently,Un-MAppable Reads Solution (UMARS) is a user-friendly web service focusing on retrieving valuable information from sequence reads that cannot be mapped back to reference genomes. Recently, next-generation sequencing (NGS) technology has emerged as a powerful tool for generating high-throughput sequencing data and has been applied to many kinds of biological research. In a typical analysis, adaptor-trimmed NGS reads were first mapped back to reference sequences, including genomes or transcripts. However, a fraction of NGS reads failed to be mapped back to the reference sequences. Such un-mappable reads are usually imputed to sequencing errors and discarded without further consideration. We are investigating possible biological relevance and possible sources of un-mappable reads. Therefore, we developed UMARS to scan for virus genomic fragments or exon-exon junctions of novel alternative splicing isoforms from un-mappable reads. For mapping un-mappable reads, we first collected viral genomes and sequences of exon-exon junctions. Then, we constructed UMARS pipeline as an automatic alignment interface. By demonstrating the results of two UMARS alignment cases, we show the applicability of UMARS. We first showed that the expected EBV genomic fragments can be detected by UMARS. Second, we also detected exon-exon junctions from un-mappable reads. Further experimental validation also ensured the authenticity of the UMARS pipeline. The UMARS service is freely available to the academic community and can be accessed via http://musk.ibms.sinica.edu.tw/UMARS/. In this study, we have shown that some un-mappable reads are not caused by sequencing errors. They can originate from viral infection or transcript splicing. Our UMARS pipeline provides another way to examine and recycle the un-mappable reads that are commonly discarded as garbage.

• Gene expression-based chemical genomics identifies potential therapeutic drugs in hepatocellular carcinoma.

  Authors: Ming-Huang Chen, Wu-Lung R Yang, Kuan-Ting Lin, Chia-Hung Liu, Yu-Wen Liu, Kai-Wen Huang, Peter Mu-Hsin Chang, Jin-Mei Lai, Chun-Nan Hsu, Kun-Mao Chao, Cheng-Yan Kao, Chi-Ying F Huang

  PloS one. 01/2011; 6(11):e27186.

  Hepatocellular carcinoma (HCC) is an aggressive tumor with a poor prognosis. Currently, only sorafenib is approved by the FDA for advanced HCC treatment; therefore, there is an urgent need toHepatocellular carcinoma (HCC) is an aggressive tumor with a poor prognosis. Currently, only sorafenib is approved by the FDA for advanced HCC treatment; therefore, there is an urgent need to discover candidate therapeutic drugs for HCC. We hypothesized that if a drug signature could reverse, at least in part, the gene expression signature of HCC, it might have the potential to inhibit HCC-related pathways and thereby treat HCC. To test this hypothesis, we first built an integrative platform, the "Encyclopedia of Hepatocellular Carcinoma genes Online 2", dubbed EHCO2, to systematically collect, organize and compare the publicly available data from HCC studies. The resulting collection includes a total of 4,020 genes. To systematically query the Connectivity Map (CMap), which includes 6,100 drug-mediated expression profiles, we further designed various gene signature selection and enrichment methods, including a randomization technique, majority vote, and clique analysis. Subsequently, 28 out of 50 prioritized drugs, including tanespimycin, trichostatin A, thioguanosine, and several anti-psychotic drugs with anti-tumor activities, were validated via MTT cell viability assays and clonogenic assays in HCC cell lines. To accelerate their future clinical use, possibly through drug-repurposing, we selected two well-established drugs to test in mice, chlorpromazine and trifluoperazine. Both drugs inhibited orthotopic liver tumor growth. In conclusion, we successfully discovered and validated existing drugs for potential HCC therapeutic use with the pipeline of Connectivity Map analysis and lab verification, thereby suggesting the usefulness of this procedure to accelerate drug repurposing for HCC treatment.

• From NPC therapeutic target identification to potential treatment strategy.

  Authors: Ming-Ying Lan, Chi-Long Chen, Kuan-Ting Lin, Sheng-An Lee, Wu-Lung R Yang, Chun-Nan Hsu, Jaw-Ching Wu, Ching-Yin Ho, Jin-Ching Lin, Chi-Ying F Huang

  Molecular cancer therapeutics. 09/2010; 9(9):2511-23.

  Nasopharyngeal carcinoma (NPC) is relatively rare in Western countries but is a common cancer in southern Asia. Many differentially expressed genes have been linked to NPC; however, how to prioritizeNasopharyngeal carcinoma (NPC) is relatively rare in Western countries but is a common cancer in southern Asia. Many differentially expressed genes have been linked to NPC; however, how to prioritize therapeutic targets and potential drugs from unsorted gene lists remains largely unknown. We first collected 558 upregulated and 993 downregulated NPC genes from published microarray data and the primary literatures. We then postulated that conversion of gene signatures into the protein-protein interaction network and analyzing the network topologically could provide insight into key regulators involved in tumorigenesis of NPC. Of particular interest was the presence of cliques, called fully connected subgraphs, in the inferred NPC networks. These clique-based hubs, connecting with more than three queries and ranked higher than other nodes in the NPC protein-protein interaction network, were further narrowed down by pathway analysis to retrieve 24 upregulated and 6 downregulated bottleneck genes for predicting NPC carcinogenesis. Moreover, additional oncogenes, tumor suppressor genes, genes involved in protein complexes, and genes obtained after functional profiling were merged with the bottleneck genes to form the final gene signature of 38 upregulated and 10 downregulated genes. We used the initial and final NPC gene signatures to query the Connectivity Map, respectively, and found that target reduction through our pipeline could efficiently uncover potential drugs with cytotoxicity to NPC cancer cells. An integrative Web site (http://140.109.23.188:8080/NPC) was established to facilitate future NPC research. This in silico approach, from target prioritization to potential drugs identification, might be an effective method for various cancer researches.

• A spectral graph theoretic approach to quantification and calibration of collective morphological differences in cell images.

  Authors: Yu-Shi Lin, Chung-Chih Lin, Yuh-Show Tsai, Tien-Chuan Ku, Yi-Hung Huang, Chun-Nan Hsu

  Bioinformatics (Oxford, England). 06/2010; 26(12):i29-37.

  High-throughput image-based assay technologies can rapidly produce a large number of cell images for drug screening, but data analysis is still a major bottleneck that limits their utility.High-throughput image-based assay technologies can rapidly produce a large number of cell images for drug screening, but data analysis is still a major bottleneck that limits their utility. Quantifying a wide variety of morphological differences observed in cell images under different drug influences is still a challenging task because the result can be highly sensitive to sampling and noise. We propose a graph-based approach to cell image analysis. We define graph transition energy to quantify morphological differences between image sets. A spectral graph theoretic regularization is applied to transform the feature space based on training examples of extremely different images to calibrate the quantification. Calibration is essential for a practical quantification method because we need to measure the confidence of the quantification. We applied our method to quantify the degree of partial fragmentation of mitochondria in collections of fluorescent cell images. We show that with transformation, the quantification can be more accurate and sensitive than that without transformation. We also show that our method outperforms competing methods, including neighbourhood component analysis and the multi-variate drug profiling method by Loo et al. We illustrate its utility with a study of Annonaceous acetogenins, a family of compounds with drug potential. Our result reveals that squamocin induces more fragmented mitochondria than muricin A. Mitochondrial cell images, their corresponding feature sets (SSLF and WSLF) and the source code of our proposed method are available at http://aiia.iis.sinica.edu.tw/. Supplementary data are available at Bioinformatics online.

• Identification of homologous microRNAs in 56 animal genomes.

  Authors: Sung-Chou Li, Wen-Ching Chan, Ling-Yueh Hu, Chun-Hung Lai, Chun-Nan Hsu, Wen-chang Lin

  Genomics. 03/2010; 96(1):1-9.

  MicroRNAs (miRNAs) are endogenous non-protein-coding RNAs of approximately 22 nucleotides. Thousands of miRNA genes have been identified (computationally and/or experimentally) in a variety ofMicroRNAs (miRNAs) are endogenous non-protein-coding RNAs of approximately 22 nucleotides. Thousands of miRNA genes have been identified (computationally and/or experimentally) in a variety of organisms, which suggests that miRNA genes have been widely shared and distributed among species. Here, we used unique miRNA sequence patterns to scan the genome sequences of 56 bilaterian animal species for locating candidate miRNAs first. The regions centered surrounding these candidate miRNAs were then extracted for folding and calculating the features of their secondary structure. Using a support vector machine (SVM) as a classifier combined with these features, we identified an additional 13,091 orthologous or paralogous candidate pre-miRNAs, as well as their corresponding candidate mature miRNAs. Stem-loop RT-PCR and deep sequencing methods were used to experimentally validate the prediction results in human, medaka and rabbit. Our prediction pipeline allows the rapid and effective discovery of homologous miRNAs in a large number of genomes.

• Fusion of systems for automated cell phenotype image classification.

  Authors: Loris Nanni, Alessandra Lumini, Yu-Shi Lin, Chun-Nan Hsu, Chung-Chih Lin

  Expert Syst. Appl. 01/2010; 37:1556-1562.

• Learning to predict expression efficacy of vectors in recombinant protein production.

  Authors: Wen-Ching Chan, Po-Huang Liang, Yan-Ping Shih, Ueng-Cheng Yang, Wen-chang Lin, Chun-Nan Hsu

  BMC Bioinformatics. 01/2010; 11:21.

• Discovery and characterization of medaka miRNA genes by next generation sequencing platform.

  Authors: Sung-Chou Li, Wen-Ching Chan, Meng-Ru Ho, Kuo-Wang Tsai, Ling-Yueh Hu, Chun-Hung Lai, Chun-Nan Hsu, Pung-Pung Hwang, Wen-chang Lin

  BMC genomics. 01/2010; 11 Suppl 4:S8.

  MicroRNAs (miRNAs) are endogenous non-protein-coding RNA genes which exist in a wide variety of organisms, including animals, plants, virus and even unicellular organisms. Medaka (Oryzias latipes) isMicroRNAs (miRNAs) are endogenous non-protein-coding RNA genes which exist in a wide variety of organisms, including animals, plants, virus and even unicellular organisms. Medaka (Oryzias latipes) is a useful model organism among vertebrate animals. However, no medaka miRNAs have been investigated systematically. It is beneficial to conduct a genome-wide miRNA discovery study using the next generation sequencing (NGS) technology, which has emerged as a powerful sequencing tool for high-throughput analysis. In this study, we adopted ABI SOLiD platform to generate small RNA sequence reads from medaka tissues, followed by mapping these sequence reads back to medaka genome. The mapped genomic loci were considered as candidate miRNAs and further processed by a support vector machine (SVM) classifier. As result, we identified 599 novel medaka pre-miRNAs, many of which were found to encode more than one isomiRs. Besides, additional minor miRNAs (also called miRNA star) can be also detected with the improvement of sequencing depth. These quantifiable isomiRs and minor miRNAs enable us to further characterize medaka miRNA genes in many aspects. First of all, many medaka candidate pre-miRNAs position close to each other, forming many miRNA clusters, some of which are also conserved across other vertebrate animals. Secondly, during miRNA maturation, there is an arm selection preference of mature miRNAs within precursors. We observed the differences on arm selection preference between our candidate pre-miRNAs and their orthologous ones. We classified these differences into three categories based on the distribution of NGS reads. Finally, we also investigated the relationship between conservation status and expression level of miRNA genes. We concluded that the evolutionally conserved miRNAs were usually the most abundant ones. Medaka is a widely used model animal and usually involved in many biomedical studies, including the ones on development biology. Identifying and characterizing medaka miRNA genes would benefit the studies using medaka as a model organism.

• Sequence features involved in the mechanism of 3' splice junction wobbling.

  Authors: Kuo-Wang Tsai, Wen-Ching Chan, Chun-Nan Hsu, Wen-Chang Lin

  BMC molecular biology. 01/2010; 11:34.

  Alternative splicing is an important mechanism mediating the diversified functions of genes in multicellular organisms, and such event occurs in around 40-60% of human genes. Recently, a newAlternative splicing is an important mechanism mediating the diversified functions of genes in multicellular organisms, and such event occurs in around 40-60% of human genes. Recently, a new splice-junction wobbling mechanism was proposed that subtle modifications exist in mRNA maturation by alternatively choosing at 5'- GTNGT and 3'- NAGNAG, which created single amino acid insertion and deletion isoforms. By browsing the Alternative Splicing Database information, we observed that most 3' alternative splice site choices occur within six nucleotides of the dominant splice site and the incidence significantly decreases further away from the dominant acceptor site. Although a lower frequency of alternative splicing occurs within the intronic region (alternative splicing at the proximal AG) than in the exonic region (alternative splicing at the distal AG), alternative AG sites located within the intronic region show stronger potential as the acceptor. These observations revealed that the choice of 3' splice sites during 3' splicing junction wobbling could depend on the distance between the duplicated AG and the branch point site (BPS). Further mutagenesis experiments demonstrated that the distance of AG-to-AG and BPS-to-AG can greatly influence 3' splice site selection. Knocking down a known alternative splicing regulator, hSlu7, failed to affect wobble splicing choices. Our results implied that nucleotide distance between proximal and distal AG sites has an important regulatory function. In this study, we showed that occurrence of 3' wobble splicing occurs in a distance-dependent manner and that most of this wobble splicing is probably caused by steric hindrance from a factor bound at the neighboring tandem motif sequence.

• Learning to predict expression efficacy of vectors in recombinant protein production.

  Authors: Wen-Ching Chan, Po-Huang Liang, Yan-Ping Shih, Ueng-Cheng Yang, Wen-chang Lin, Chun-Nan Hsu

  BMC bioinformatics. 01/2010; 11 Suppl 1:S21.

  Recombinant protein production is a useful biotechnology to produce a large quantity of highly soluble proteins. Currently, the most widely used production system is to fuse a target protein intoRecombinant protein production is a useful biotechnology to produce a large quantity of highly soluble proteins. Currently, the most widely used production system is to fuse a target protein into different vectors in Escherichia coli (E. coli). However, the production efficacy of different vectors varies for different target proteins. Trial-and-error is still the common practice to find out the efficacy of a vector for a given target protein. Previous studies are limited in that they assumed that proteins would be over-expressed and focused only on the solubility of expressed proteins. In fact, many pairings of vectors and proteins result in no expression. In this study, we applied machine learning to train prediction models to predict whether a pairing of vector-protein will express or not express in E. coli. For expressed cases, the models further predict whether the expressed proteins would be soluble. We collected a set of real cases from the clients of our recombinant protein production core facility, where six different vectors were designed and studied. This set of cases is used in both training and evaluation of our models. We evaluate three different models based on the support vector machines (SVM) and their ensembles. Unlike many previous works, these models consider the sequence of the target protein as well as the sequence of the whole fusion vector as the features. We show that a model that classifies a case into one of the three classes (no expression, inclusion body and soluble) outperforms a model that considers the nested structure of the three classes, while a model that can take advantage of the hierarchical structure of the three classes performs slight worse but comparably to the best model. Meanwhile, compared to previous works, we show that the prediction accuracy of our best method still performs the best. Lastly, we briefly present two methods to use the trained model in the design of the recombinant protein production systems to improve the chance of high soluble protein production. In this paper, we show that a machine learning approach to the prediction of the efficacy of a vector for a target protein in a recombinant protein production system is promising and may compliment traditional knowledge-driven study of the efficacy. We will release our program to share with other labs in the public domain when this paper is published.

• Feature Space Transformation for Semi-Supervised Learning for Protein Subcellular Localization in Fluorescence Microscopy Images.

  Authors: Yu-Shi Lin, Yi-Hung Huang, Chung-Chih Lin, Chun-Nan Hsu

  Proceedings of the 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Boston, MA, USA, June 28 - July 1, 2009; 01/2009

• Periodic Step Size Adaptation for Single Pass On-line Learning.

  Authors: Chun-Nan Hsu, Yu-Ming Chang, Han-Shen Huang, Yuh-Jye Lee

  Advances in Neural Information Processing Systems 22: 23rd Annual Conference on Neural Information Processing Systems 2009. Proceedings of a meeting held 7-10 December 2009, Vancouver, British Columbia, Canada.; 01/2009

• BIOADI: a machine learning approach to identifying abbreviations and definitions in biological literature.

  Authors: Cheng-Ju Kuo, Maurice H. T. Ling, Kuan-Ting Lin, Chun-Nan Hsu

  BMC Bioinformatics. 01/2009; 10:7.

• Global and componentwise extrapolations for accelerating training of Bayesian networks and conditional random fields.

  Authors: Han-Shen Huang, Bo-Hou Yang, Yu-Ming Chang, Chun-Nan Hsu

  Data Min. Knowl. Discov. 01/2009; 19:58-94.

• BIOADI: a machine learning approach to identifying abbreviations and definitions in biological literature

  Authors: Cheng-Ju Kuo, Maurice Ling, Kuan-Ting Lin, Chun-Nan Hsu

  BMC Bioinformatics. 01/2009;

  Abstract Background To automatically process large quantities of biological literature for knowledge discovery and information curation, text mining tools are becoming essential. AbbreviationAbstract Background To automatically process large quantities of biological literature for knowledge discovery and information curation, text mining tools are becoming essential. Abbreviation recognition is related to NER and can be considered as a pair recognition task of a terminology and its corresponding abbreviation from free text. The successful identification of abbreviation and its corresponding definition is not only a prerequisite to index terms of text databases to produce articles of related interests, but also a building block to improve existing gene mention tagging and gene normalization tools. Results Our approach to abbreviation recognition (AR) is based on machine-learning, which exploits a novel set of rich features to learn rules from training data. Tested on the AB3P corpus, our system demonstrated a F-score of 89.90% with 95.86% precision at 84.64% recall, higher than the result achieved by the existing best AR performance system. We also annotated a new corpus of 1200 PubMed abstracts which was derived from BioCreative II gene normalization corpus. On our annotated corpus, our system achieved a F-score of 86.20% with 93.52% precision at 79.95% recall, which also outperforms all tested systems. Conclusion By applying our system to extract all short form-long form pairs from all available PubMed abstracts, we have constructed BIOADI. Mining BIOADI reveals many interesting trends of bio-medical research. Besides, we also provide an off-line AR software in the download section on http://bioagent.iis.sinica.edu.tw/BIOADI/ .

• Periodic step-size adaptation in second-order gradient descent for single-pass on-line structured learning.

  Authors: Chun-Nan Hsu, Han-Shen Huang, Yu-Ming Chang, Yuh-Jye Lee

  Machine Learning. 01/2009; 77:195-224.