Jingwen Yan

Alzheimer's disease (AD) is the most common neurodegenerative disorder. Using discrete disease status as the phenotype and computing statistics at the single marker level may not be able to address the underlying biological interactions that contribute to disease mechanism and may contribute to the issue of "missing heritability." We performed a genome-wide association study (GWAS) and a genome-wide interaction study (GWIS) of an amyloid imaging phenotype, using the data from Alzheimer's Disease Neuroimaging Initiative. We investigated the genetic main effects and interaction effects on cingulate amyloid-beta (Aβ) load in an effort to better understand the genetic etiology of Aβ deposition that is a widely studied AD biomarker. PLINK was used in the single marker GWAS, and INTERSNP was used to perform the two-marker GWIS, focusing only on SNPs with p ≤ 0.01 for the GWAS analysis. Age, sex, and diagnosis were used as covariates in both analyses. Corrected p values u...

As the most common type of dementia, Alzheimer's disease (AD) is a neurodegenerative disorder initially manifested by impaired memory performances. While the diagnosis information indicates a dichotomous status of a patient, memory scores have the potential to capture the continuous nature of the disease progression and may provide more insights into the underlying mechanism. In this work, we performed a targeted genetic study of memory scores on an AD cohort to identify the associations between a set of genes highly expressed in the hippocampal region and seven cognitive scores related to episodic memory. Both main effects and interaction effects of the targeted genetic markers on these correlated memory scores were examined. In addition to well-known AD genetic markers APOE and TOMM40, our analysis identified a new risk gene NAV2 through the gene-level main effect analysis. NAV2 was found to be significantly and consistently associated with all seven episodic memory scores. Ge...

Brain imaging genetics is an emergent research field where the association between genetic variations such as single nucleotide polymorphisms (SNPs) and neuroimaging quantitative traits (QTs) is evaluated. Sparse canonical correlation analysis (SCCA) is a bi-multivariate analysis method that has the potential to reveal complex multi-SNP-multi-QT associations. We present initial efforts on evaluating a few SCCA methods for brain imaging genetics. This includes a data synthesis method to create realistic imaging genetics data with known SNP-QT associations, application of three SCCA algorithms to the synthetic data, and comparative study of their performances. Our empirical results suggest, approximating covariance structure using an identity or diagonal matrix, an approach used in these SCCA algorithms, could limit the SCCA capability in identifying the underlying imaging genetics associations. An interesting future direction is to develop enhanced SCCA methods that effectively take into account the covariance structures in the imaging genetics data.

Regression models have been widely studied to investigate the prediction power of neuroimaging measures as biomarkers for inferring cognitive outcomes in the Alzheimer's disease study. Most of these models ignore the interrelated structures either within neuroimaging measures or between cognitive outcomes, and thus may have limited power to yield optimal solutions. To address this issue, we propose to use a new sparse multitask learning model called Group-Sparse Multi-task Regression and Feature Selection (G-SMuRFS) and demonstrate its effectiveness by examining the predictive power of detailed cortical thickness measures toward 3 types of cognitive scores in a large cohort. G-SMuRFS proposes a group-level l2,1-norm strategy to group relevant features together in an anatomically meaningful manner and use this prior knowledge to guide the learning process. This approach also takes into account the correlation among cognitive outcomes for building a more appropriate predictive mod...

Imaging genetics is an emerging field that studies the influence of genetic variation on brain structure and function. The major task is to examine the association between genetic markers such as single-nucleotide polymorphisms (SNPs) and quantitative traits (QTs) extracted from neuroimaging data. The complexity of these datasets has presented critical bioinformatics challenges that require new enabling tools. Sparse canonical correlation analysis (SCCA) is a bi-multivariate technique used in imaging genetics to identify complex multi-SNP-multi-QT associations. However, most of the existing SCCA algorithms are designed using the soft thresholding method, which assumes that the input features are independent from one another. This assumption clearly does not hold for the imaging genetic data. In this article, we propose a new knowledge-guided SCCA algorithm (KG-SCCA) to overcome this limitation as well as improve learning results by incorporating valuable prior knowledge. The propose...

Identification of biomarkers for early detection of Alzheimer's disease (AD) is an important research topic. Prior work has shown that multimodal imaging and biomarker data could provide complementary information for prediction of cognitive or AD status. However, the relationship among multiple data modalities are often ignored or oversimplified in prior studies. To address this issue, we propose a network-guided sparse learning model to embrace the complementary information and inter-relationships between modalities. We apply this model to predict cognitive outcome from imaging and proteomic data, and show that the proposed model not only outperforms traditional ones, but also yields stable multimodal biomarkers across cross-validation trials.

ABSTRACT Regression models have been widely studied to investigate whether multimodal neuroimaging measures can be used as effective biomarkers for predicting cognitive outcomes in the study of Alzheimer's Disease (AD). Most existing models overlook the interrelated structures either within neuroimaging measures or between cognitive outcomes, and thus may have limited power to yield optimal solutions. To address this issue, we propose to incorporate an ℓ2,1 norm and/or a group ℓ2,1 norm (G2,1 norm) in the regression models. Using ADNI-1 and ADNI-GO/2 data, we apply these models to examining the ability of structural MRI and AV-45 PET scans for predicting cognitive measures including ADAS and RAVLT scores. We focus our analyses on the participants with mild cognitive impairment (MCI), a prodromal stage of AD, in order to identify useful patterns for early detection. Compared with traditional linear and ridge regression methods, these new models not only demonstrate superior and more stable predictive performances, but also identify a small set of imaging markers that are biologically meaningful.

ABSTRACT There has been strong current interest to manipulate collective spins and even induce magnetic phase transitions in their highly non-equilibrium, non-thermal states at femtosecond time scales. Such processes offer opportunities to exceed the upper limit of the magnetic switching speed (0.1-10 GHz) in modern magneto-optical recording industry and magnetic storage/logic devices. One prominent system to explore such femtosecond magnetism is strongly correlated manganites, which are truly "responsive" near the phase boundary, exhibiting extreme sensitivity to external stimuli, such as light, electric and magnetic fields. Using ultrafast two-color magnetic circular dichroism spectroscopy, we have observed a substantial photoinduced magnetization enhancement in Pr0.7Ca0.3MnO3 within 180 fs above a threshold pump fluence and at low temperature. Such a photoinduced critical behavior vanishes at elevated temperature. These results clearly show a photoinduced ultrafast antiferromagnetic to ferromagnetic phase transition, demonstrating particularly, that one can reveal a hidden, thermally inaccessible ground state at fs time scales.

Alzheimer's disease (AD) is characterized by gradual neurodegeneration and loss of brain function, especially for memory during early stages. Regression analysis has been widely applied to AD research to relate clinical and biomarker data such as predicting cognitive outcomes from MRI measures. In particular, sparse models have been proposed to identify the optimal imaging markers with high prediction power. However, the complex relationship among imaging markers are often overlooked or simplified in the existing methods. To address this issue, we present a new sparse learning method by introducing a novel network term to more flexibly model the relationship among imaging markers. The proposed algorithm is applied to the ADNI study for predicting cognitive outcomes using MRI scans. The effectiveness of our method is demonstrated by its improved prediction performance over several state-of-the-art competing methods and accurate identification of cognition-relevant imaging markers that are biologically meaningful.

Brain imaging genetics is an emergent research field where the association between genetic variations such as single nucleotide polymorphisms (SNPs) and neuroimaging quantitative traits (QTs) is evaluated. Sparse canonical correlation analysis (SCCA) is a bi-multivariate analysis method that has the potential to reveal complex multi-SNP-multi-QT associations. Most existing SCCA algorithms are designed using the soft threshold strategy, which assumes that the features in the data are independent from each other. This independence assumption usually does not hold in imaging genetic data, and thus inevitably limits the capability of yielding optimal solutions. We propose a novel structure-aware SCCA (denoted as S2CCA) algorithm to not only eliminate the independence assumption for the input data, but also incorporate group-like structure in the model. Empirical comparison with a widely used SCCA implementation, on both simulated and real imaging genetic data, demonstrated that S2CCA could yield improved prediction performance and biologically meaningful findings.

ABSTRACT Object extraction from still images is an important task in pattern recognition and computer vision. It is very hard to find a fully automatic object extraction method in practical applications. Therefore, a good solution is to interactively extract objects from complex background with a few simple user inputs. This paper presents an interactive object extraction method from still images. A still image is divided into many small regions by some low level segmentation methods at first. Strokes (or markers), which could be as simple as a few points, are manually input by the user to indicate the initial three types of regions, i.e., object regions, background regions and unmarked regions. A region is merged with its adjacent region if (1) both of the regions are of the same type or (2) either of them is an unmarked region and the similarity between them is among the k-global maximal ones. The region merging process completes when each initial unmarked region is identified as either object or background and the similarity between any two regions equals zero. The extensive experiments demonstrate the efficiency and effectiveness of the proposed method. (c) 2013 Elsevier B.V. All rights reserved.