Tsung Fei Khang

Agreement analysis of LENSTAR with other techniques of biometry

PurposeTo assess the agreement of the optical low-coherence reflectometry (OLCR) device LENSTAR LS900 with partial coherence interferometry (PCI) device IOLMaster and applanation and immersion ultrasound biometry.MethodsWe conducted the study at the Ophthalmology Clinic, University of Malaya Medical Center, Malaysia. Phakic eyes of 76 consecutive cataract patients were measured using four different methods: IOLMaster, LENSTAR and A-scan applanation and immersion ultrasound biometry. We assessed the method agreement in the LENSTAR-IOLMaster, LENSTAR-applanation, and LENSTAR-immersion comparisons for axial length (AL) and intraocular lens (IOL) power using Bland–Altman plots. For average K, we compared LENSTAR with IOLMaster and the TOPCON KR-8100 autorefractor-keratometer. SRK/T formula was used to compute IOL power, with emmetropia as the target refractive outcome.ResultsFor all the variables studied, LENSTAR agreement with IOLMaster is strongest, followed by those with immersion and applanation. For the LENSTAR-IOLMaster comparison, the estimated proportion of differences falling within 0.33 mm from zero AL and within 1D from zero IOL power is 100%. The estimated proportion of differences falling within 0.5 D from zero average K is almost 100% in the LENSTAR-IOLMaster comparison but 88% in the LENSTAR-TOPCON comparison. The proportion of differences falling within 0.10 mm (AL) and within 1D (IOL power) in the LENSTAR-IOLMaster comparison has practically significant discrepancy with that of LENSTAR-applanation and LENSTAR-immersion comparisons.ConclusionsIn phakic eyes of cataract patients, measurements of AL, average K, and IOL power calculated using the SRK/T formula from LENSTAR are biometrically equivalent to those from IOLMaster, but not with those from applanation and immersion ultrasound biometry.

A phylogenomic approach to bacterial subspecies classification: proof of concept in Mycobacterium abscessus.

BackgroundMycobacterium abscessus is a rapidly growing mycobacterium that is often associated with human infections. The taxonomy of this species has undergone several revisions and is still being debated. In this study, we sequenced the genomes of 12 M. abscessus strains and used phylogenomic analysis to perform subspecies classification.ResultsA data mining approach was used to rank and select informative genes based on the relative entropy metric for the construction of a phylogenetic tree. The resulting tree topology was similar to that generated using the concatenation of five classical housekeeping genes: rpoB, hsp65, secA, recA and sodA. Additional support for the reliability of the subspecies classification came from the analysis of erm41 and ITS gene sequences, single nucleotide polymorphisms (SNPs)-based classification and strain clustering demonstrated by a variable number tandem repeat (VNTR) assay and a multilocus sequence analysis (MLSA). We subsequently found that the concatenation of a minimal set of three median-ranked genes: DNA polymerase III subunit alpha (polC), 4-hydroxy-2-ketovalerate aldolase (Hoa) and cell division protein FtsZ (ftsZ), is sufficient to recover the same tree topology. PCR assays designed specifically for these genes showed that all three genes could be amplified in the reference strain of M. abscessus ATCC 19977T.ConclusionThis study provides proof of concept that whole-genome sequence-based data mining approach can provide confirmatory evidence of the phylogenetic informativeness of existing markers, as well as lead to the discovery of a more economical and informative set of markers that produces similar subspecies classification in M. abscessus. The systematic procedure used in this study to choose the informative minimal set of gene markers can potentially be applied to species or subspecies classification of other bacteria.

A New Generalization of the Logarithmic Distribution Arising from the Inverse Trinomial Distribution

A new generalization of the logarithmic distribution (LD)—the generalized logarithmic distribution (GLD)—is introduced. The GLD is a cluster size distribution of the inverse trinomial distribution, which arises in a generalized random walk problem. Major properties of the GLD are derived and discussed. The method of first moment and frequency of one count and maximum likelihood estimation (MLE) are used for parameter estimation. Raos score test statistic is derived as a model selection tool. Finally, we demonstrate an application of the GLD in fitting insect data from temperate and tropical regions.

Monogenean anchor morphometry: systematic value, phylogenetic signal, and evolution

Background. Anchors are one of the important attachment appendages for monogenean parasites. Common descent and evolutionary processes have left their mark on anchor morphometry, in the form of patterns of shape and size variation useful for systematic and evolutionary studies. When combined with morphological and molecular data, analysis of anchor morphometry can potentially answer a wide range of biological questions. Materials and Methods. We used data from anchor morphometry, body size and morphology of 13 Ligophorus (Monogenea: Ancyrocephalidae) species infecting two marine mugilid (Teleostei: Mugilidae) fish hosts: Moolgarda buchanani (Bleeker) and Liza subviridis (Valenciennes) from Malaysia. Anchor shape and size data (n = 530) were generated using methods of geometric morphometrics. We used 28S rRNA, 18S rRNA, and ITS1 sequence data to infer a maximum likelihood phylogeny. We discriminated species using principal component and cluster analysis of shape data. Adams’s Kmult was used to detect phylogenetic signal in anchor shape. Phylogeny-correlated size and shape changes were investigated using continuous character mapping and directional statistics, respectively. We assessed morphological constraints in anchor morphometry using phylogenetic regression of anchor shape against body size and anchor size. Anchor morphological integration was studied using partial least squares method. The association between copulatory organ morphology and anchor shape and size in phylomorphospace was used to test the Rohde-Hobbs hypothesis. We created monogeneaGM, a new R package that integrates analyses of monogenean anchor geometric morphometric data with morphological and phylogenetic data. Results. We discriminated 12 of the 13 Ligophorus species using anchor shape data. Significant phylogenetic signal was detected in anchor shape. Thus, we discovered new morphological characters based on anchor shaft shape, the length between the inner root point and the outer root point, and the length between the inner root point and the dent point. The species on M. buchanani evolved larger, more robust anchors; those on L. subviridis evolved smaller, more delicate anchors. Anchor shape and size were significantly correlated, suggesting constraints in anchor evolution. Tight integration between the root and the point compartments within anchors confirms the anchor as a single, fully integrated module. The correlation between male copulatory organ morphology and size with anchor shape was consistent with predictions from the Rohde-Hobbs hypothesis. Conclusions. Monogenean anchors are tightly integrated structures, and their shape variation correlates strongly with phylogeny, thus underscoring their value for systematic and evolutionary biology studies. Our MonogeneaGM R package provides tools for researchers to mine biological insights from geometric morphometric data of speciose monogenean genera.

The structural effect of intravitreal Brilliant blue G and Indocyanine green in rats eyes

PurposeTo compare the potential retinal toxicity of two commercially Brilliant blue G dyes (Brilliant Peel and Ocublue Plus) and Indocyanine green (ICG) at usual clinical concentration.MethodsBrilliant Peel 0.025% (n=9), Ocublue Plus 0.025% (n=9), and ICG 0.05% (n=9) were injected intravitreally into Sprague–Dawley rat left eyes with balanced salt solution injected in the contralateral eyes as control. Evaluation of the effect of the dyes on retinal architecture was done by histological analysis of neurosensory retinal thickness and retinal ganglion cell (RGC) counts 7 days after intravitreal injection. Paired t-test was done to detect the presence of biologically significant thinning in neurosensory retina and five retinal layers for each dye (paired t-tests). One-way ANOVA and Tukey’s Honestly Significant Difference test were used to assess whether different dyes caused significant thinning in mean neurosensory retinal thickness and reduction of mean RGC density.ResultsEyes treated with ICG had significantly thinner mean total neurosensory retinal thickness compared with the control eyes (P-value=0.01), followed by those treated with Ocublue Plus (P-value=0.03). Brilliant Peel did not cause significant thinning in any of the five retinal layers (all P-values>0.05). No significant difference in mean thinning of the total retinal thickness was detected between dyes (P-value=0.11). The mean thickness of the photoreceptor outer segment and outer plexiform layers were significantly reduced in ICG-injected eyes when compared with the control eyes (P-value=0.02). No significant difference in mean thinning between the three dyes was detected at all five retinal layers using one-way ANOVA (all P-values>0.35). RGC density was significantly reduced for ICG (P-value=0.01) but only marginally for Ocublue Plus (P-value=0.05). No significant reduction in RGC density was observed for Brilliant Peel (P-value=0.2).ConclusionIntravitreal Brilliant Peel is safe to rats retina. The retinal thinning and reduction in RGC density induced by Ocublue Plus requires further studies to determine the safety profile of this product. Potential retinal toxicity is seen with ICG 0.05%.

Getting the most out of RNA-seq data analysis

Background. A common research goal in transcriptome projects is to find genes that are differentially expressed in different phenotype classes. Biologists might wish to validate such gene candidates experimentally, or use them for downstream systems biology analysis. Producing a coherent differential gene expression analysis from RNA-seq count data requires an understanding of how numerous sources of variation such as the replicate size, the hypothesized biological effect size, and the specific method for making differential expression calls interact. We believe an explicit demonstration of such interactions in real RNA-seq data sets is of practical interest to biologists. Results. Using two large public RNA-seq data sets—one representing strong, and another mild, biological effect size—we simulated different replicate size scenarios, and tested the performance of several commonly-used methods for calling differentially expressed genes in each of them. We found that, when biological effect size was mild, RNA-seq experiments should focus on experimental validation of differentially expressed gene candidates. Importantly, at least triplicates must be used, and the differentially expressed genes should be called using methods with high positive predictive value (PPV), such as NOISeq or GFOLD. In contrast, when biological effect size was strong, differentially expressed genes mined from unreplicated experiments using NOISeq, ASC and GFOLD had between 30 to 50% mean PPV, an increase of more than 30-fold compared to the cases of mild biological effect size. Among methods with good PPV performance, having triplicates or more substantially improved mean PPV to over 90% for GFOLD, 60% for DESeq2, 50% for NOISeq, and 30% for edgeR. At a replicate size of six, we found DESeq2 and edgeR to be reasonable methods for calling differentially expressed genes at systems level analysis, as their PPV and sensitivity trade-off were superior to the other methods’. Conclusion. When biological effect size is weak, systems level investigation is not possible using RNAseq data, and no meaningful result can be obtained in unreplicated experiments. Nonetheless, NOISeq or GFOLD may yield limited numbers of gene candidates with good validation potential, when triplicates or more are available. When biological effect size is strong, NOISeq and GFOLD are effective tools for detecting differentially expressed genes in unreplicated RNA-seq experiments for qPCR validation. When triplicates or more are available, GFOLD is a sharp tool for identifying high confidence differentially expressed genes for targeted qPCR validation; for downstream systems level analysis, combined results from DESeq2 and edgeR are useful.

Comparative analysis of viral genomes from acute and chronic hepatitis B reveals novel variants associated with a lower rate of chronicity

Infection with the hepatitis B virus (HBV) may lead to an acute or chronic infection. It is generally accepted that the clinical outcome of infection depends on the balance between host immunity and viral survival strategies. In order to persist, the virus needs to have a high rate of replication and some immune‐escape capabilities. Hence, HBVs lacking these properties are likely to be eliminated more rapidly by the host, leading to a lower rate of chronicity. To test this hypothesis, 177 HBV genomes from acute non‐fulminant cases and 1,149 from chronic cases were retrieved from GenBank for comparative analysis. Selection of candidate nucleotides associated with the disease state was done using random guess cut‐off and the Bonferroni correction. Five significant nucleotides were detected using this filtering step. Their predictive values were assessed using the support vector machine classification with five‐fold cross‐validation. The average prediction accuracy was 61% ± 1%, with a sensitivity of 24% ± 1%, specificity of 98% ± 1%, positive predictive value of 92% ± 4% and negative predictive value of 56% ± 1%. BCP/X, enhancer I and surface/polymerase variants were found to be associated almost exclusively with acute hepatitis. These HBV variants are novel potential markers for non‐progression to chronic hepatitis. J. Med. Virol. 85:419–424, 2013.

CORNAS: coverage-dependent RNA-Seq analysis of gene expression data without biological replicates

BackgroundIn current statistical methods for calling differentially expressed genes in RNA-Seq experiments, the assumption is that an adjusted observed gene count represents an unknown true gene count. This adjustment usually consists of a normalization step to account for heterogeneous sample library sizes, and then the resulting normalized gene counts are used as input for parametric or non-parametric differential gene expression tests. A distribution of true gene counts, each with a different probability, can result in the same observed gene count. Importantly, sequencing coverage information is currently not explicitly incorporated into any of the statistical models used for RNA-Seq analysis.ResultsWe developed a fast Bayesian method which uses the sequencing coverage information determined from the concentration of an RNA sample to estimate the posterior distribution of a true gene count. Our method has better or comparable performance compared to NOISeq and GFOLD, according to the results from simulations and experiments with real unreplicated data. We incorporated a previously unused sequencing coverage parameter into a procedure for differential gene expression analysis with RNA-Seq data.ConclusionsOur results suggest that our method can be used to overcome analytical bottlenecks in experiments with limited number of replicates and low sequencing coverage. The method is implemented in CORNAS (Coverage-dependent RNA-Seq), and is available at https://github.com/joel-lzb/CORNAS.

An age-adjusted seroprevalence study of Toxoplasma antibody in a Malaysian ophthalmology unit.

Toxoplasma gondii is a public health risk in developing countries, especially those located in the tropics. Widespread infection may inflict a substantial burden on state resources, as patients can develop severe neurological defects and ocular diseases that result in lifelong loss of economic independence. We tested sera for IgG antibody from 493 eye patients in Malaysia. Overall age-adjusted seroprevalence was estimated to be 25% (95% CI: [21%, 29%]). We found approximately equal age-adjusted seroprevalence in Chinese (31%; 95% CI: [25%, 38%]) and Malays (29%; 95% CI: [21%, 36%]), followed by Indians (19%; 95% CI: [13%, 25%]). A logistic regression of the odds for T. gondii seroprevalence against age, gender, ethnicity and the occurrence of six types of ocular diseases showed that only age and ethnicity were significant predictors. The odds for T. gondii seroprevalence were 2.7 (95% CI for OR: [1.9, 4.0]) times higher for a patient twice as old as the other, with ethnicity held constant. In Malays, we estimated the odds for T. gondii seroprevalence to be 2.9 (95% CI for OR: [1.8, 4.5]) times higher compared to non-Malays, with age held constant. Previous studies of T. gondii seroprevalence in Malaysia did not explicitly adjust for age, rendering comparisons difficult. Our study highlights the need to adopt a more rigorous epidemiological approach in monitoring T. gondii seroprevalence in Malaysia.

GENIPAC: A Genomic Information Portal for Head and Neck Cancer Cell Systems

Head and neck cancer (HNC)–derived cell lines represent fundamental models for studying the biological mechanisms underlying cancer development and precision therapies. However, mining the genomic information of HNC cells from available databases requires knowledge on bioinformatics and computational skill sets. Here, we developed a user-friendly web resource for exploring, visualizing, and analyzing genomics information of commonly used HNC cell lines. We populated the current version of GENIPAC with 44 HNC cell lines from 3 studies: ORL Series, OPC-22, and H Series. Specifically, the mRNA expressions for all the 3 studies were derived with RNA-seq. The copy number alterations analysis of ORL Series was performed on the Genome Wide Human Cytoscan HD array, while copy number alterations for OPC-22 were derived from whole exome sequencing. Mutations from ORL Series and H Series were derived from RNA-seq information, while OPC-22 was based on whole exome sequencing. All genomic information was preprocessed with customized scripts and underwent data validation and correction through data set validator tools provided by cBioPortal. The clinical and genomic information of 44 HNC cell lines are easily assessable in GENIPAC. The functional utility of GENIPAC was demonstrated with some of the genomic alterations that are commonly reported in HNC, such as TP53, EGFR, CCND1, and PIK3CA. We showed that these genomic alterations as reported in The Cancer Genome Atlas database were recapitulated in the HNC cell lines in GENIPAC. Importantly, genomic alterations within pathways could be simultaneously visualized. We developed GENIPAC to create access to genomic information on HNC cell lines. This cancer omics initiative will help the research community to accelerate better understanding of HNC and the development of new precision therapeutic options for HNC treatment. GENIPAC is freely available at http://genipac.cancerresearch.my/.