Identifying gene set association enrichment using the coefficient of intrinsic dependence

Identification of DYNLT1 associated with proliferation, relapse, and metastasis in breast cancer

Background

Breast cancer (BC) is the most common malignant disease worldwide. Although the survival rate is improved in recent years, the prognosis is still bleak once recurrence and metastasis occur. It is vital to investigate more efficient biomarkers for predicting the metastasis and relapse of BC. DYNLT1 has been reported that participating in the progression of multiple cancers. However, there is still a lack of study about the correlation between DYNLT1 and BC.

Methods

In this study, we evaluated and validated the expression pattern and prognostic implication of DYNLT1 in BC with multiple public cohorts and BC tumor microarrays (TMAs) of paraffin-embedded tissues collected from the Affiliated Hospital of Jining Medical University. The response biomarkers for immune therapy, such as tumor mutational burden (TMB), between different DYNLT1 expression level BC samples were investigated using data from the TCGA-BRCA cohort utilizing public online tools. In addition, colony formation and transwell assay were conducted to verify the effects of DYNLT1 in BC cell line proliferation and invasion.

Results

The results demonstrated that DYNLT1 overexpressed in BC and predicted poor relapse-free survival in our own BC TMA cohort. In addition, DYNLT1 induced BC development by promoting MDA-MB-231 cell proliferation migration, and metastasis.

Conclusion

Altogether, our findings proposed that DYNLT1 could be a diagnostic and prognostic indicator in BC.

Redundancy-aware unsupervised ranking based on game theory: Ranking pathways in collections of gene sets

In Genetics, gene sets are grouped in collections concerning their biological function. This often leads to high-dimensional, overlapping, and redundant families of sets, thus precluding a straightforward interpretation of their biological meaning. In Data Mining, it is often argued that techniques to reduce the dimensionality of data could increase the maneuverability and consequently the interpretability of large data. In the past years, moreover, we witnessed an increasing consciousness of the importance of understanding data and interpretable models in the machine learning and bioinformatics communities. On the one hand, there exist techniques aiming to aggregate overlapping gene sets to create larger pathways. While these methods could partly solve the large size of the collections' problem, modifying biological pathways is hardly justifiable in this biological context. On the other hand, the representation methods to increase interpretability of collections of gene sets that have been proposed so far have proved to be insufficient. Inspired by this Bioinformatics context, we propose a method to rank sets within a family of sets based on the distribution of the singletons and their size. We obtain sets' importance scores by computing Shapley values; Making use of microarray games, we do not incur the typical exponential computational complexity. Moreover, we address the challenge of constructing redundancy-aware rankings where, in our case, redundancy is a quantity proportional to the size of intersections among the sets in the collections. We use the obtained rankings to reduce the dimension of the families, therefore showing lower redundancy among sets while still preserving a high coverage of their elements. We finally evaluate our approach for collections of gene sets and apply Gene Sets Enrichment Analysis techniques to the now smaller collections: As expected, the unsupervised nature of the proposed rankings allows for unremarkable differences in the number of significant gene sets for specific phenotypic traits. In contrast, the number of performed statistical tests can be drastically reduced. The proposed rankings show a practical utility in bioinformatics to increase interpretability of the collections of gene sets and a step forward to include redundancy-awareness into Shapley values computations.

Construction of gene causal regulatory networks using microarray data with the coefficient of intrinsic dependence

BACKGROUND

In the past two decades, biologists have been able to identify the gene signatures associated with various phenotypes through the monitoring of gene expressions with high-throughput biotechnologies. These gene signatures have in turn been successfully applied to drug development, disease prevention, crop improvement, etc. However, ignoring the interactions among genes has weakened the predictive power of gene signatures in practical applications. Gene regulatory networks, in which genes are represented by nodes and the associations between genes are represented by edges, are typically constructed to analyze and visualize such gene interactions. More specifically, the present study sought to measure gene-gene associations by using the coefficient of intrinsic dependence (CID) to capture more nonlinear as well as cause-effect gene relationships.

RESULTS

A stepwise procedure using the CID along with the partial coefficient of intrinsic dependence (pCID) was demonstrated for the rebuilding of simulated networks and the well-known CBF-COR pathway under cold stress using Arabidopsis microarray data. The procedure was also applied to the construction of bHLH gene regulatory pathways under abiotic stresses using rice microarray data, in which OsbHLH104, a putative phytochrome-interacting factor (OsPIF14), and OsbHLH060, a positive regulator of iron homeostasis (OsPRI1) were inferred as the most affiliated genes. The inferred regulatory pathways were verified through literature reviews.

CONCLUSIONS

The proposed method can efficiently decipher gene regulatory pathways and may assist in achieving higher predictive power in practical applications. The lack of any mention in the literature of some of the regulatory event may have been due to the high complexity of the regulatory systems in the plant transcription, a possibility which could potentially be confirmed in the near future given ongoing rapid developments in bio-technology.

Microarray meta-analysis to explore abiotic stress-specific gene expression patterns in Arabidopsis

BACKGROUND

Abiotic stresses are the major limiting factors that affect plant growth, development, yield and final quality. Deciphering the underlying mechanisms of plants' adaptations to stresses using few datasets might overlook the different aspects of stress tolerance in plants, which might be simultaneously and consequently operated in the system. Fortunately, the accumulated microarray expression data offer an opportunity to infer abiotic stress-specific gene expression patterns through meta-analysis. In this study, we propose to combine microarray gene expression data under control, cold, drought, heat, and salt conditions and determined modules (gene sets) of genes highly associated with each other according to the observed expression data.

RESULTS

By analyzing the expression variations of the Eigen genes from different conditions, we had identified two, three, and five gene modules as cold-, heat-, and salt-specific modules, respectively. Most of the cold- or heat-specific modules were differentially expressed to a particular degree in shoot samples, while most of the salt-specific modules were differentially expressed to a particular degree in root samples. A gene ontology (GO) analysis on the stress-specific modules suggested that the gene modules exclusively enriched stress-related GO terms and that different genes under the same GO terms may be alternatively disturbed in different conditions. The gene regulatory events for two genes, DREB1A and DEAR1, in the cold-specific gene module had also been validated, as evidenced through the literature search.

CONCLUSIONS

Our protocols study the specificity of the gene modules that were specifically activated under a particular type of abiotic stress. The biplot can also assist to visualize the stress-specific gene modules. In conclusion, our approach has the potential to further elucidate mechanisms in plants and beneficial for future experiments design under different abiotic stresses.

A Stepwise Approach of Finding Dependent Variables via Coefficient of Intrinsic Dependence

The coefficient of intrinsic dependence (CID) is capable of determining associations among variables without making distributional or functional assumptions regarding random variables. In this study, we developed the partial coefficient of intrinsic dependence (pCID) to facilitate the step-by-step selection of variables that are relevant to a target variable. The strategy of selecting relevant variables using the CID along with the pCID can eliminate interference from other relevant variables. From simulation results, we observed that the proposed method is more sensitive to curvilinearity and more specific to linearity than the combination of Pearsons correlation coefficient and the partial correlation coefficient (PCC/pPCC). This property may provide the opportunity to index different levels of curvilinearity according to CID/pCID outcomes. In practice trials conducted using publicly available microarray data, the CID/pCID procedure successfully identified cold-responsive genes related to three C-repeat binding factors, and was especially effective at identifying some sample-specific gene-gene interactions. Therefore, the proposed strategy may be beneficial in meta-analysis to distinguish general forms of relationships from the noise.