NID2 and HOXA9 promoter hypermethylation as biomarkers for prevention and early detection in oral cavity squamous cell carcinoma tissues and saliva

Differentially methylated oral squamous cell carcinoma (OSCC) biomarkers, identified in vitro and validated in well-characterized surgical specimens, have shown poor clinical correlation in cohorts with different risk profiles. To overcome this lack of relevance, we used the HumanMethylation27 BeadChip, publicly available methylation and expression array data, and quantitative methylation specific PCR to uncover differential methylation in OSCC clinical samples with heterogeneous risk profiles. A two stage design consisting of discovery and prevalence screens was used to identify differential promoter methylation and deregulated pathways in patients diagnosed with OSCC and head and neck squamous cell carcinoma. Promoter methylation of KIF1A (κ = 0.64), HOXA9 (κ = 0.60), NID2 (κ = 0.60), and EDNRB (κ = 0.60) had a moderate to substantial agreement with clinical diagnosis in the discovery screen. HOXA9 had 68% sensitivity, 100% specificity, and a 0.81 Area Under the Curve (AUC). NID2 had 71% sensitivity, 100% specificity, and a 0.79 AUC. In the prevalence screen, HOXA9 (κ = 0.82) and NID2 (κ = 0.80) had an almost perfect agreement with histologic diagnosis. HOXA9 had 85% sensitivity, 97% specificity, and a 0.95 AUC. NID2 had 87% sensitivity, 95% specificity, and a 0.91 AUC. A HOXA9 and NID2 gene panel had 94% sensitivity, 97% specificity, and a 0.97 AUC. In saliva, from OSCC cases and controls, HOXA9 had 75% sensitivity, 53% specificity, and a 0.75 AUC. NID2 had 87% sensitivity, 21% specificity, and a 0.73 AUC. This phase I Biomarker Development Trial identified a panel of differentially methylated genes in normal and OSCC clinical samples from patients with heterogeneous risk profiles. This panel may be useful for early detection and cancer prevention studies.

Epigenetic memory in induced pluripotent stem cells

Somatic cell nuclear transfer and transcription-factor-based reprogramming revert adult cells to an embryonic state, and yield pluripotent stem cells that can generate all tissues. Through different mechanisms and kinetics, these two reprogramming methods reset genomic methylation, an epigenetic modification of DNA that influences gene expression, leading us to hypothesize that the resulting pluripotent stem cells might have different properties. Here we observe that low-passage induced pluripotent stem cells (iPSCs) derived by factor-based reprogramming of adult murine tissues harbour residual DNA methylation signatures characteristic of their somatic tissue of origin, which favours their differentiation along lineages related to the donor cell, while restricting alternative cell fates. Such an 'epigenetic memory' of the donor tissue could be reset by differentiation and serial reprogramming, or by treatment of iPSCs with chromatin-modifying drugs. In contrast, the differentiation and methylation of nuclear-transfer-derived pluripotent stem cells were more similar to classical embryonic stem cells than were iPSCs. Our data indicate that nuclear transfer is more effective at establishing the ground state of pluripotency than factor-based reprogramming, which can leave an epigenetic memory of the tissue of origin that may influence efforts at directed differentiation for applications in disease modelling or treatment.

Estimation and assessment of raw copy numbers at the single locus level

MOTIVATION

Although copy-number aberrations are known to contribute to the diversity of the human DNA and cause various diseases, many aberrations and their phenotypes are still to be explored. The recent development of single-nucleotide polymorphism (SNP) arrays provides researchers with tools for calling genotypes and identifying chromosomal aberrations at an order-of-magnitude greater resolution than possible a few years ago. The fundamental problem in array-based copy-number (CN) analysis is to obtain CN estimates at a single-locus resolution with high accuracy and precision such that downstream segmentation methods are more likely to succeed.

RESULTS

We propose a preprocessing method for estimating raw CNs from Affymetrix SNP arrays. Its core utilizes a multichip probe-level model analogous to that for high-density oligonucleotide expression arrays. We extend this model by adding an adjustment for sequence-specific allelic imbalances such as cross-hybridization between allele A and allele B probes. We focus on total CN estimates, which allows us to further constrain the probe-level model to increase the signal-to-noise ratio of CN estimates. Further improvement is obtained by controlling for PCR effects. Each part of the model is fitted robustly. The performance is assessed by quantifying how well raw CNs alone differentiate between one and two copies on Chromosome X (ChrX) at a single-locus resolution (27kb) up to a 200kb resolution. The evaluation is done with publicly available HapMap data.

AVAILABILITY

The proposed method is available as part of an open-source R package named aroma.affymetrix. Because it is a bounded-memory algorithm, any number of arrays can be analyzed.

On the utility of pooling biological samples in microarray experiments

Over 15% of the data sets catalogued in the Gene Expression Omnibus Database involve RNA samples that have been pooled before hybridization. Pooling affects data quality and inference, but the exact effects are not yet known because pooling has not been systematically studied in the context of microarray experiments. Here we report on the results of an experiment designed to evaluate the utility of pooling and the impact on identifying differentially expressed genes. We find that inference for most genes is not adversely affected by pooling, and we recommend that pooling be done when fewer than three arrays are used in each condition. For larger designs, pooling does not significantly improve inferences if few subjects are pooled. The realized benefits in this case do not outweigh the price paid for loss of individual specific information. Pooling is beneficial when many subjects are pooled, provided that independent samples contribute to multiple pools.

A comparison of normalization methods for high density oligonucleotide array data based on variance and bias

MOTIVATION

When running experiments that involve multiple high density oligonucleotide arrays, it is important to remove sources of variation between arrays of non-biological origin. Normalization is a process for reducing this variation. It is common to see non-linear relations between arrays and the standard normalization provided by Affymetrix does not perform well in these situations.

RESULTS

We present three methods of performing normalization at the probe intensity level. These methods are called complete data methods because they make use of data from all arrays in an experiment to form the normalizing relation. These algorithms are compared to two methods that make use of a baseline array: a one number scaling based algorithm and a method that uses a non-linear normalizing relation by comparing the variability and bias of an expression measure. Two publicly available datasets are used to carry out the comparisons. The simplest and quickest complete data method is found to perform favorably.

AVAILABILITY

Software implementing all three of the complete data normalization methods is available as part of the R package Affy, which is a part of the Bioconductor project http://www.bioconductor.org.

SUPPLEMENTARY INFORMATION

Additional figures may be found at http://www.stat.berkeley.edu/~bolstad/normalize/index.html

Electrocorticographic gamma activity during word production in spoken and sign language

OBJECTIVE

To investigate the functional-neuroanatomic substrates of word production using signed versus spoken language.

METHODS

The authors studied single-word processing with varying input and output modalities in a 38-year-old woman with normal hearing and speech who had become proficient in sign language 8 years before developing intractable epilepsy. Subdural electrocorticography (ECoG) was performed during picture naming and word reading (visual inputs) and word repetition (auditory inputs); these tasks were repeated with speech and with sign language responses. Cortical activation was indexed by event-related power augmentation in the 80- to 100-Hz gamma band, and was compared with general principles of functional anatomy and with subject-specific maps of the same or similar tasks using electrical cortical stimulation (ECS).

RESULTS

Speech outputs activated tongue regions of the sensorimotor cortex, and sign outputs activated hand regions. In addition, signed word production activated parietal regions that were not activated by spoken word production. Posterior superior temporal gyrus was activated earliest and to the greatest extent during auditory word repetition, and the basal temporal-occipital cortex was activated similarly during naming and reading, reflecting the different modalities of input processing. With few exceptions, topographic patterns of ECoG gamma were consistent with ECS maps of the same or similar language tasks.

CONCLUSIONS

Spoken and signed word production activated many of the same cortical regions, particularly those processing auditory and visual inputs; however, they activated different regions of sensorimotor cortex, and signing activated parietal cortex more than did speech. This study illustrates the utility of electrocorticographic gamma for studying the neuroanatomy and processing dynamics of human language.

Assessing homeostasis through circadian patterns

An organism is thought to be in a dynamic state of homeostasis when each physiological and behavioral system reaches a delicate balance within the framework of other regulatory processes. Many biological systems target specific set-point variables and generate circadian patterns. In this article, we focus on specific measurements representative of two systems, namely deep-body temperature and activity counts. We examine data collected every 30 minutes in mice, assume there are underlying circadian patterns, and extend the approach presented in Brumback and Rice (1998, Journal of the American Statistical Association 93, 961-976) in order to obtain estimates in the presence of correlated data. We then assess homeostasis using these estimates and their statistical properties.

Cross-correlation of fetal cardiac and somatic activity as an indicator of antenatal neural development

OBJECTIVE

In this study, we wanted to model the emergence of coupling between fetal cardiac and somatic activity in normal and at-risk fetuses.

STUDY DESIGN

One hundred six fetuses of uncomplicated pregnancies were longitudinally monitored at 20, 24, 28, 32, 36, and 38 weeks of gestation by using a fetal actocardiograph and computerized data collection. Twenty-six fetuses of complicated pregnancies were also included. Statistical time series analysis techniques were used to examine the relation between fetal movement and fetal heart rate.

RESULTS

A linear increase was found in the magnitude of the cross-correlation function between fetal movement and fetal heart rate as gestation advanced, with coalescence around a peak lag of 5 seconds by 32 weeks. Fetuses that delivered before term evidenced accelerated fetal movement and fetal heart rate coupling, whereas fetuses affected by deleterious conditions showed a decline in developmental trajectory.

CONCLUSIONS

The cross-correlation between fetal cardiac and somatic activity is an indicator of neuroregulation in human fetuses.