PubMed Health⌕ Search

Biomedical subjects

Hongzhe Li

Publications and source records attributed to Hongzhe Li.

At least 19 recordsLinked to original sources

Persistent antimicrobial resistance during soil remediation driven by residual heavy metal co-selection.

Remediation of heavy metal-contaminated soil is a global priority, particularly as reclaimed land increasingly intersects with urban development and human exposure. However, the ecological consequences of soil remediation, especially its impact on antimicrobial resistance (AMR) as a global health threat, have remained poorly understood. Here, we combined single-cell Raman-D₂O probing with genome-resolved metagenomics to monitor the dynamics of phenotypic and genotypic resistance to metals and antibiotics during a 120-day remediation of soils with three contamination levels from a lead-zinc smelting site. Although chemical remediation substantially reduced bioavailable metals (by 42%-65%), AMR was not diminished. Instead, both phenotypic activity and gene abundance of metal- and antibiotic-resistant microorganisms increased, resulting in a two- to three-fold increase in AMR-associated health risks. Among 76 metagenome assembled genomes (MAGs) from phenotypic resistance communities, all Cd resistance-associated MAGs harbored multidrug resistance genes, half of which were colocalized with metal resistance determinants, and their prevalence continued to rise with remediation. These findings reveal that although remediation alleviates acute metal toxicity, residual low-concentration bioavailable metals sustain evolutionary selection for resistance, highlighting a disconnect between chemical recovery and biological safety. Moreover, the improved soil nutrient and physiochemical properties of remediated soils further promoted the proliferation of antibiotic-resistant bacteria. This study offers new ecological insights into the unintended consequences of anthropogenic interventions, underscoring the need to integrate biological safety into soil health and safety assessments.

Soil Microbiology↗

dGAMLSS: an exact, distributed algorithm to fit Generalized Additive Models for Location, Scale, and Shape for privacy-preserving population reference charts.

MOTIVATION: There is growing interest in estimating population reference ranges across age and sex to better identify atypical clinically-relevant measurements throughout the lifespan. For this task, the World Health Organization recommends using Generalized Additive Models for Location, Scale, and Shape (GAMLSS), which can model non-linear growth trajectories under complex distributions that address the heterogeneity in human populations.Fitting GAMLSS models requires large, generalizable sample sizes, especially for accurate estimation of extreme quantiles, but obtaining such multi-site data can be challenging due to privacy concerns and practical considerations. In settings where patient data cannot be shared, privacy-preserving distributed algorithms for federated learning can be used, but no such algorithm exists for GAMLSS. RESULTS: We propose distributed GAMLSS (dGAMLSS), a distributed algorithm that can fit GAMLSS models across multiple sites without sharing patient-level data. This includes specific considerations for the fitting of smooth functions at varying levels of communication efficiency. We demonstrate the effectiveness of dGAMLSS in constructing population reference charts across clinical, genomics, and neuroimaging settings and show that dGAMLSS is able to reproduce pooled reference charts and inference down to numerical differences. AVAILABILITY AND IMPLEMENTATION: An R package providing examples of the dGAMLSS algorithm, as well as functions for sharing and aggregating site-specific parameters, is available at https://github.com/hufengling/dGAMLSS.

Algorithms↗

SirT1 modulates the estrogen-insulin-like growth factor-1 signaling for postnatal development of mammary gland in mice.

INTRODUCTION: Estrogen and insulin-like growth factor-1 (IGF-1) play important roles in mammary gland development and breast cancer. SirT1 is a highly conserved protein deacetylase that can regulate the insulin/IGF-1 signaling in lower organisms, as well as a growing number of transcription factors, including NF-kappaB, in mammalian cells. Whether SirT1 regulates the IGF-1 signaling for mammary gland development and function, however, is not clear. In the present study, this role of SirT1 was examined by studying SirT1-deficient mice. METHODS: SirT1-deficient (SirT1(ko/ko)) mice were generated by crossing a new strain of mice harboring a conditional targeted mutation in the SirT1 gene (SirT1(co/co)) with CMV-Cre transgenic mice. Whole mount and histology analyses, immunofluorescence staining, immunohistochemistry, and western blotting were used to characterize mammary gland development in virgin and pregnant mice. The effect of exogenous estrogen was also examined by subcutaneous implantation of a slow-releasing pellet in the subscapular region. RESULTS: Both male and female SirT1(ko/ko) mice can be fertile despite the growth retardation phenotype. Virgin SirT1(ko/ko) mice displayed impeded ductal morphogenesis, whereas pregnant SirT1(ko/ko) mice manifested lactation failure due to an underdeveloped lobuloalveolar network. Estrogen implantation was sufficient to rescue ductal morphogenesis. Exogenous estrogen reversed the increased basal level of IGF-1 binding protein-1 expression in SirT1(ko/ko) mammary tissues, but not that of IkappaB alpha expression, suggesting that increased levels of estrogen enhanced the production of local IGF-1 and rescued ductal morphogenesis. Additionally, TNFalpha treatment enhanced the level of the newly synthesized IkappaB alpha in SirT1(ko/ko) cells. SirT1 deficiency therefore affects the cellular response to multiple extrinsic signals. CONCLUSION: SirT1 modulates the IGF-1 signaling critical for both growth regulation and mammary gland development in mice. SirT1 deficiency deregulates the expression of IGF-1 binding protein-1 and attenuates the effect of IGF-1 signals, including estrogen-stimulated local IGF-1 signaling for the onset of ductal morphogenesis. These findings suggest that the enzymatic activity of SirT1 may influence both normal growth and malignant growth of mammary epithelial cells.

Animals↗

Promotion of cellular migration and apoptosis resistance by a mouse eye-specific phosphatidylethanolamine-binding protein.

The phosphatidylethanolamine (PE)-binding protein (PEBP) family is an evolutionary conserved group of proteins found in various species with multiple functions. We identified human PE-binding protein 4 (hPEBP4), a novel member of the PEBP family, as an anti-apoptotic molecule. Here we describe the cloning and functional characterization of the mouse homolog of hPEBP4 (mouse PEBP4, mPEBP4). Full-length cDNA of mPEBP4 contains 1048 bp with an open reading frame (ORF) of 729 bp, which is predicted to encode a 242-aa protein. mPEBP4 contains a PE-binding domain, in this case between amino acids 106 and 213. mPEBP4 localizes primarily to endoplasmic reticulum/Golgi apparatus. Interestingly, RT-PCR and in situ hybridization analyses indicate that mPEBP4 is specifically expressed in mouse eye tissue. We demonstrate that mPEBP4 promotes cellular migration and invasion by inhibiting ERK1/2 and JNK activation and up-regulating the expression of COX-2. In addition, mPEBP4 overexpression inhibits Epirubicin-induced cellular apoptosis. Considering that mPEBP4 is specifically expressed in retinal ganglion cells, whether mPEBP4 is an important molecule involved in visual function needs to be further investigated.

Amino Acid Sequence↗

hPEBP4 resists TRAIL-induced apoptosis of human prostate cancer cells by activating Akt and deactivating ERK1/2 pathways.

The treatment options available for prostate cancer are limited because of its resistance to therapeutic agents. Thus, a better understanding of the underlying mechanisms of the resistance of prostate cancer will facilitate the discovery of more efficient treatment protocols. Human phosphatidylethanolamine-binding protein 4 (hPEBP4) is recently identified by us as an anti-apoptotic molecule and a potential candidate target for breast cancer treatment. Here we found the expression levels of hPEBP4 were positively correlated with the severity of clinical prostate cancer. Furthermore, hPEBP4 was not expressed in TRAIL-sensitive DU145 prostate cancer cells, but was highly expressed in TRAIL-resistant LNCaP cells, which show highly activated Akt. Interestingly, hPEBP4 overexpression in TRAIL-sensitive DU145 cells promoted Akt activation but inhibited ERK1/2 activation. The hPEBP4-overexpressing DU145 cells became resistant to TRAIL-induced apoptosis consequently, which could be reversed by PI3K inhibitors. In contrast, silencing of hPEBP4 in TRAIL-resistant LNCaP cells inhibited Akt activation but increased ERK1/2 activation, resulting in their sensitivity to TRAIL-induced apoptosis that was restored by the MEK1 inhibitor. Therefore, hPEBP4 expression in prostate cancer can activate Akt and deactivate ERK1/2 signaling, leading to TRAIL resistance. We also demonstrated that hPEBP4-mediated resistance to TRAIL-induced apoptosis occurred downstream of caspase-8 and at the level of BID cleavage via the regulation of Akt and ERK pathways, and that hPEBP4-regulated ERK deactivation was upstream of Akt activation in prostate cancer cells. Considering that hPEBP4 confers cellular resistance to TRAIL-induced apoptosis and is abundantly expressed in poorly differentiated prostate cancer, silencing of hPEBP4 suggests a promising approach for prostate cancer treatment.

Apoptosis↗

Responses of inferior colliculus neurons to SAM tones located in inhibitory response areas.

In order to examine the effect of inhibition on processing auditory temporal information, responses of single neurons in the inferior colliculus of the chinchilla to sinusoidally amplitude-modulated (SAM) tones alone and the presence of a steady-state tone were obtained. The carrier frequency of the SAM tone was either the characteristic frequency (CF) or a frequency in the inhibitory response area of a studied neuron. When the carrier frequency was set to the neuron's CF, neurons responded in synchrony to the SAM-tone envelope, as expected. When the carrier frequency was set to a frequency at which pure tones produced inhibition, SAM tones elicited little or no response, also as expected. However, when the same SAM tone was paired with a pure tone whose frequency was set to the neuron's CF, responses synchronized to the SAM tone envelope were obtained. These modulated responses were typically one-half cycle out-of-phase with the response to the SAM tone at CF, suggesting that they arose from cyclic inhibition and release from inhibition by the SAM tone. The results demonstrate that the representation of temporal information by inferior colliculus neurons is influenced by temporally-patterned inhibition arising from locations remote from CF.

Acoustic Stimulation↗

Nonparametric pathway-based regression models for analysis of genomic data.

High-throughout genomic data provide an opportunity for identifying pathways and genes that are related to various clinical phenotypes. Besides these genomic data, another valuable source of data is the biological knowledge about genes and pathways that might be related to the phenotypes of many complex diseases. Databases of such knowledge are often called the metadata. In microarray data analysis, such metadata are currently explored in post hoc ways by gene set enrichment analysis but have hardly been utilized in the modeling step. We propose to develop and evaluate a pathway-based gradient descent boosting procedure for nonparametric pathways-based regression (NPR) analysis to efficiently integrate genomic data and metadata. Such NPR models consider multiple pathways simultaneously and allow complex interactions among genes within the pathways and can be applied to identify pathways and genes that are related to variations of the phenotypes. These methods also provide an alternative to mediating the problem of a large number of potential interactions by limiting analysis to biologically plausible interactions between genes in related pathways. Our simulation studies indicate that the proposed boosting procedure can indeed identify relevant pathways. Application to a gene expression data set on breast cancer distant metastasis identified that Wnt, apoptosis, and cell cycle-regulated pathways are more likely related to the risk of distant metastasis among lymph-node-negative breast cancer patients. Results from analysis of other two breast cancer gene expression data sets indicate that the pathways of Metalloendopeptidases (MMPs) and MMP inhibitors, as well as cell proliferation, cell growth, and maintenance are important to breast cancer relapse and survival. We also observed that by incorporating the pathway information, we achieved better prediction for cancer recurrence.

Adult↗

Antitumor activity of a small-molecule inhibitor of human silent information regulator 2 enzymes.

SIRT1 and other NAD-dependent deacetylases have been implicated in control of cellular responses to stress and in tumorigenesis through deacetylation of important regulatory proteins, including p53 and the BCL6 oncoprotein. Hereby, we describe the identification of a compound we named cambinol that inhibits NAD-dependent deacetylase activity of human SIRT1 and SIRT2. Consistent with the role of SIRT1 in promoting cell survival during stress, inhibition of SIRT1 activity with cambinol during genotoxic stress leads to hyperacetylation of key stress response proteins and promotes cell cycle arrest. Treatment of BCL6-expressing Burkitt lymphoma cells with cambinol as a single agent induced apoptosis, which was accompanied by hyperacetylation of BCL6 and p53. Because acetylation inactivates BCL6 and has the opposite effect on the function of p53 and other checkpoint pathways, the antitumor activity of cambinol in Burkitt lymphoma cells may be accomplished through a combined effect of BCL6 inactivation and checkpoint activation. Cambinol was well tolerated in mice and inhibited growth of Burkitt lymphoma xenografts. Inhibitors of NAD-dependent deacetylases may constitute novel anticancer agents.

Acetylation↗

Self-assembled supramolecular hydrogels formed by biodegradable PEO-PHB-PEO triblock copolymers and alpha-cyclodextrin for controlled drug delivery.

A materials design of a new supramolecular hydrogel self-assembled between alpha-cyclodextrin and a biodegradable poly(ethylene oxide)-poly[(r)-3-hydroxybutyrate]-poly(ethylene oxide) (PEO-PHB-PEO) triblock copolymer was demonstrated. The cooperation effect of complexation of PEO segments with alpha-cyclodextrin and the hydrophobic interaction between PHB blocks resulted in the formation of the supramolecular hydrogel with a strong macromolecular network. The in vitro release kinetics studies of fluorescein isothiocyanate labeled dextran (dextran-FITC) model drug from the hydrogel showed that the hydrogel was suitable for relatively long-term sustained controlled release of macromolecular drugs, which many simple triblock copolymer hydrogel systems could not achieve. The hydrogel was found to be thixotropic and reversible, and can be applied as a promising injectable drug delivery system.

Biocompatible Materials↗

Gradient directed regularization for sparse Gaussian concentration graphs, with applications to inference of genetic networks.

Large-scale microarray gene expression data provide the possibility of constructing genetic networks or biological pathways. Gaussian graphical models have been suggested to provide an effective method for constructing such genetic networks. However, most of the available methods for constructing Gaussian graphs do not account for the sparsity of the networks and are computationally more demanding or infeasible, especially in the settings of high dimension and low sample size. We introduce a threshold gradient descent (TGD) regularization procedure for estimating the sparse precision matrix in the setting of Gaussian graphical models and demonstrate its application to identifying genetic networks. Such a procedure is computationally feasible and can easily incorporate prior biological knowledge about the network structure. Simulation results indicate that the proposed method yields a better estimate of the precision matrix than the procedures that fail to account for the sparsity of the graphs. We also present the results on inference of a gene network for isoprenoid biosynthesis in Arabidopsis thaliana. These results demonstrate that the proposed procedure can indeed identify biologically meaningful genetic networks based on microarray gene expression data.

Arabidopsis↗

Silencing of human phosphatidylethanolamine-binding protein 4 sensitizes breast cancer cells to tumor necrosis factor-alpha-induced apoptosis and cell growth arrest.

PURPOSE: The current therapeutic approach is not so effective in breast cancer patients. Alternative treatment protocols aimed at different targets need to be explored. We recently reported a novel phosphatidylethanolamine-binding protein, human phosphatidylethanolamine-binding protein 4 (hPEBP4), as an antiapoptotic molecule. The finding led us to explore a promising approach for breast cancer therapy via silencing the expression of hPEBP4. EXPERIMENTAL DESIGN: hPEBP4 expression in clinical breast specimens was examined by Tissue Microarrays. RNA interference was used to silence hPEBP4 expression in MCF-7 breast carcinoma cells and the effects on cell proliferation, cell cycle progression, apoptosis, as well as underlying mechanisms, were investigated. RESULTS: hPEBP4 was found to be expressed in up to 50% of breast cancers but in only <4% of normal breast tissues. Silencing of hPEBP4 potentiated tumor necrosis factor-alpha (TNF-alpha)-induced apoptosis and cell cycle arrest in MCF-7 cells, which was due to the increased mitogen-activated protein kinase activation and the enhanced phosphatidylethanolamine externalization. Further investigation showed that silencing of hPEBP4 in MCF-7 cells promoted TNF-alpha-induced stability of p53, up-regulation of phospho-p53ser15, p21waf/cip, and Bax, and down-regulation of Bcl-2 and Bcl-xL, which were shown to depend on extracellular signal-regulated kinase 1/2 and c-jun NH2-terminal kinase activation by hPEBP4 silencing. Moreover, the increased proportion of cells in the G0-G1 phase of cell cycle was observed in hPEBP4-silenced MCF-7 cells on TNF-alpha treatment and the expression of cyclin A and cyclin E was down-regulated more significantly. CONCLUSIONS: The antiapoptotic effect and the preferential expression pattern in breast cancer tissues make hPEBP4 a new target for breast cancer therapy. Silencing of hPEBP4 expression may be a promising approach for the treatment of breast carcinoma.

Apoptosis↗

Examination of ancestry and ethnic affiliation using highly informative diallelic DNA markers: application to diverse and admixed populations and implications for clinical epidemiology and forensic medicine.

We and others have identified several hundred ancestry informative markers (AIMs) with large allele frequency differences between different major ancestral groups. For this study, a panel of 199 widely distributed AIMs was used to examine a diverse set of 796 DNA samples including self-identified European Americans, West Africans, East Asians, Amerindians, African Americans, Mexicans, Mexican Americans, Puerto Ricans and South Asians. Analysis using a Bayesian clustering algorithm (STRUCTURE) showed grouping of individuals with similar ethnic identity without any identifier other than the AIMs genotyping and showed admixture proportions that clearly distinguished different individuals of mixed ancestry. Additional analyses showed that, for the majority of samples, the predicted ethnic identity corresponded with the self-identified ethnicity at high probability (P > 0.99). Overall, the study demonstrates that AIMs can provide a useful adjunct to forensic medicine, pharmacogenomics and disease studies in which major ancestry or ethnic affiliation might be linked to specific outcomes.

Algorithms↗

AAA ATPases regulate membrane association of yeast oxysterol binding proteins and sterol metabolism.

The yeast genome encodes seven oxysterol binding protein homologs, Osh1p-Osh7p, which have been implicated in regulating intracellular lipid and vesicular transport. Here, we show that both Osh6p and Osh7p interact with Vps4p, a member of the AAA (ATPases associated with a variety of cellular activities) family. The coiled-coil domain of Osh7p was found to interact with Vps4p in a yeast two-hybrid screen and the interaction between Osh7p and Vps4p appears to be regulated by ergosterol. Deletion of VPS4 induced a dramatic increase in the membrane-associated pools of Osh6p and Osh7p and also caused a decrease in sterol esterification, which was suppressed by overexpression of OSH7. Lastly, overexpression of the coiled-coil domain of Osh7p (Osh7pCC) resulted in a multivesicular body sorting defect, suggesting a dominant negative role of Osh7pCC possibly through inhibiting Vps4p function. Our data suggest that a common mechanism may exist for AAA proteins to regulate the membrane association of yeast OSBP proteins and that these two protein families may function together to control subcellular lipid transport.

Adenosine Triphosphatases↗

Prevalence of stereotypical responses to mistuned complex tones in the inferior colliculus.

The human auditory system has an exceptional ability to separate competing sounds, but the neural mechanisms that underlie this ability are not understood. Responses of inferior colliculus (IC) neurons to "mistuned" complex tones were measured to investigate possible neural mechanisms for spectral segregation. A mistuned tone is a harmonic complex tone in which the frequency of one component has been changed; that component may be heard as a separate sound source, suggesting that the mistuned tone engages the same mechanisms that contribute to the segregation of natural sounds. In this study, the harmonic tone consisted of eight harmonics of 250 Hz; in the mistuned tone, the frequency of the fourth harmonic was increased by 12% (120 Hz). The mistuned tone elicited a stereotypical discharge pattern, consisting of peaks separated by about 8 ms and a response envelope modulated with a period of 100 ms, which bore little resemblance to the discharge pattern elicited by the harmonic tone or to the stimulus waveform. Similar discharge patterns were elicited from many neurons with a range of characteristic frequencies, especially from neurons that exhibited short-latency sustained responses to pure tones. In contrast, transient and long-latency neurons usually did not exhibit the stereotypical discharge pattern. The discharge pattern was generally stable when the stimulus level or component phase was varied; the major effect of these manipulations was to shift the phase of the response envelope. Simulation of IC responses with a computational model suggested that off-frequency inhibition could produce discharge patterns with these characteristics.

Acoustic Stimulation↗

Penalized Cox regression analysis in the high-dimensional and low-sample size settings, with applications to microarray gene expression data.

MOTIVATION: An important application of microarray technology is to relate gene expression profiles to various clinical phenotypes of patients. Success has been demonstrated in molecular classification of cancer in which the gene expression data serve as predictors and different types of cancer serve as a categorical outcome variable. However, there has been less research in linking gene expression profiles to the censored survival data such as patients' overall survival time or time to cancer relapse. It would be desirable to have models with good prediction accuracy and parsimony property. RESULTS: We propose to use the L(1) penalized estimation for the Cox model to select genes that are relevant to patients' survival and to build a predictive model for future prediction. The computational difficulty associated with the estimation in the high-dimensional and low-sample size settings can be efficiently solved by using the recently developed least-angle regression (LARS) method. Our simulation studies and application to real datasets on predicting survival after chemotherapy for patients with diffuse large B-cell lymphoma demonstrate that the proposed procedure, which we call the LARS-Cox procedure, can be used for identifying important genes that are related to time to death due to cancer and for building a parsimonious model for predicting the survival of future patients. The LARS-Cox regression gives better predictive performance than the L(2) penalized regression and a few other dimension-reduction based methods. CONCLUSIONS: We conclude that the proposed LARS-Cox procedure can be very useful in identifying genes relevant to survival phenotypes and in building a parsimonious predictive model that can be used for classifying future patients into clinically relevant high- and low-risk groups based on the gene expression profile and survival times of previous patients.

Biomarkers, Tumor↗

Boosting proportional hazards models using smoothing splines, with applications to high-dimensional microarray data.

MOTIVATION: An important area of research in the postgenomics era is to relate high-dimensional genetic or genomic data to various clinical phenotypes of patients. Due to large variability in time to certain clinical events among patients, studying possibly censored survival phenotypes can be more informative than treating the phenotypes as categorical variables. Due to high dimensionality and censoring, building a predictive model for time to event is more difficult than the classification/linear regression problem. We propose to develop a boosting procedure using smoothing splines for estimating the general proportional hazards models. Such a procedure can potentially be used for identifying non-linear effects of genes on the risk of developing an event. RESULTS: Our empirical simulation studies showed that the procedure can indeed recover the true functional forms of the covariates and can identify important variables that are related to the risk of an event. Results from predicting survival after chemotherapy for patients with diffuse large B-cell lymphoma demonstrate that the proposed method can be used for identifying important genes that are related to time to death due to cancer and for building a parsimonious model for predicting the survival of future patients. In addition, there is clear evidence of non-linear effects of some genes on survival time.

Algorithms↗

A novel human phosphatidylethanolamine-binding protein resists tumor necrosis factor alpha-induced apoptosis by inhibiting mitogen-activated protein kinase pathway activation and phosphatidylethanolamine externalization.

The phosphatidylethanolamine (PE)-binding proteins (PEBPs) are an evolutionarily conserved family of proteins with pivotal biological functions. Here we describe the cloning and functional characterization of a novel family member, human phosphatidylethanolamine-binding protein 4 (hPEBP4). hPEBP4 is expressed in most human tissues and highly expressed in tumor cells. Its expression in tumor cells is further enhanced upon tumor necrosis factor (TNF) alpha treatment, whereas hPEBP4 normally co-localizes with lysosomes, TNFalpha stimulation triggers its transfer to the cell membrane, where it binds to Raf-1 and MEK1. L929 cells overexpressing hPEBP4 are resistant to both TNFalpha-induced ERK1/2, MEK1, and JNK activation and TNFalpha-mediated apoptosis. Co-precipitation and in vitro protein binding assay demonstrated that hPEBP4 interacts with Raf-1 and MEK1. A truncated form of hPEBP4, lacking the PE-binding domain, maintains lysosomal co-localization but has no effect on cellular responses to TNFalpha. Given that MCF-7 breast cancer cells expressed hPEBP4 at a high level, small interfering RNA was used to silence the expression of hPEBP4. We demonstrated that down-regulation of hPEBP4 expression sensitizes MCF-7 breast cancer cells to TNFalpha-induced apoptosis. hPEBP4 appears to promote cellular resistance to TNF-induced apoptosis by inhibiting activation of the Raf-1/MEK/ERK pathway, JNK, and PE externalization, and the conserved region of PE-binding domain appears to play a vital role in this biological activity of hPEBP4.

Amino Acid Sequence↗

Partial Cox regression analysis for high-dimensional microarray gene expression data.

MOTIVATION: An important application of microarray technology is to predict various clinical phenotypes based on the gene expression profile. Success has been demonstrated in molecular classification of cancer in which different types of cancer serve as categorical outcome variable. However, there has been less research in linking gene expression profile to censored survival outcome such as patients' overall survival time or time to cancer relapse. In this paper, we develop a partial Cox regression method for constructing mutually uncorrelated components based on microarray gene expression data for predicting the survival of future patients. RESULTS: The proposed partial Cox regression method involves constructing predictive components by repeated least square fitting of residuals and Cox regression fitting. The key difference from the standard principal components of Cox regression analysis is that in constructing the predictive components, our method utilizes the observed survival/censoring information. We also propose to apply the time-dependent receiver operating characteristic curve analysis to evaluate the results. We applied our methods to a publicly available dataset of diffuse large B-cell lymphoma. The outcomes indicated that combining the partial Cox regression method with principal components analysis results in parsimonious model with fewer components and better predictive performance. We conclude that the proposed partial Cox regression method can be very useful in building a parsimonious predictive model that can accurately predict the survival of future patients based on the gene expression profile and survival times of previous patients. AVAILABILITY: R codes are available upon request.

Biomarkers, Tumor↗