LDPCD: A Novel Method for Locally Differentially Private Community Detection

As one of the cores of data analysis in large social networks, community detection has become a hot research topic in recent years. However, user’s real social relationship may be at risk of privacy leakage and threatened by inference attacks because of the semitrusted server. As a result, community detection in social graphs under local differential privacy has gradually aroused the interest of industry and academia. On the one hand, the distortion of user’s real data caused by existing privacy-preserving mechanisms can have a serious impact on the mining process of densely connected local graph structure, resulting in low utility of the final community division. On the other hand, private community detection requires to use the results of multiple user-server interactions to adjust user’s partition, which inevitably leads to excessive allocation of privacy budget and large error of perturbed data. For these reasons, a new community detection method based on the local differential privacy model (named LDPCD) is proposed in this paper. Due to the introduction of truncated Laplace mechanism, the accuracy of user perturbation data is improved. In addition, the community divisive algorithm based on extremal optimization (EO) is also refined to reduce the number of interactions between users and the server. Thus, the total privacy overhead is reduced and strong privacy protection is guaranteed. Finally, LDPCD is applied in two commonly used real-world datasets, and its advantage is experimentally validated compared with two state-of-the-art methods.

A new class of fractional impulsive differential hemivariational inequalities with an application

We consider a new fractional impulsive differential hemivariational inequality, which captures the required characteristics of both the hemivariational inequality and the fractional impulsive differential equation within the same framework. By utilizing a surjectivity theorem and a fixed point theorem we establish an existence and uniqueness theorem for such a problem. Moreover, we investigate the perturbation problem of the fractional impulsive differential hemivariational inequality to prove a convergence result, which describes the stability of the solution in relation to perturbation data. Finally, our main results are applied to obtain some new results for a frictional contact problem with the surface traction driven by the fractional impulsive differential equation.

Prediction of drug targets for specific diseases leveraging gene perturbation data: A machine learning approach

Identification of the correct targets is a key element for successful drug development. However, there are limited approaches for predicting drug targets for specific diseases using omics data, and few have leveraged expression profiles from gene perturbations. We present a novel computational target discovery approach based on machine learning (ML) models. ML models are first trained on drug-induced expression profiles, with outcomes defined as whether the drug treats the studied disease. The goal is to learn expression patterns associated with treatment. The fitted ML models were then applied to expression profiles from gene perturbations (over-expression[OE]/knockdown[KD]). We prioritized targets based on predicted probabilities from the ML model, which reflects treatment potential. The methodology was applied to predict targets for hypertension, diabetes mellitus (DM), rheumatoid arthritis (RA) and schizophrenia (SCZ). We validated our approach by evaluating whether the identified targets may re-discover known drug targets from an external database (OpenTargets). We indeed found evidence of significant enrichment across all diseases under study. Further literature search revealed that many candidates were supported by previous studies. For example, we predicted PSMB8 inhibition to be associated with treatment of RA, which was supported by a study showing PSMB8 inhibitors (PR-957) ameliorated experimental RA in mice. In conclusion, we propose a new ML approach to integrate expression profiles from drugs and gene perturbations and validated the framework. Our approach is flexible and may provide an independent source of information when prioritizing targets.

Neuroblastoma signalling models unveil combination therapies targeting feedback-mediated resistance

Very high risk neuroblastoma is characterised by increased MAPK signalling, and targeting MAPK signalling is a promising therapeutic strategy. We used a deeply characterised panel of neuroblastoma cell lines and found that the sensitivity to MEK inhibitors varied drastically between these cell lines. By generating quantitative perturbation data and mathematical modelling, we determined potential resistance mechanisms. We found that negative feedbacks within MAPK signalling and via the IGF receptor mediate re-activation of MAPK signalling upon treatment in resistant cell lines. By using cell-line specific models, we predict that combinations of MEK inhibitors with RAF or IGFR inhibitors can overcome resistance, and tested these predictions experimentally. In addition, phospho-proteomic profiling confirmed the cell-specific feedback effects and synergy of MEK and IGFR targeted treatment. Our study shows that a quantitative understanding of signalling and feedback mechanisms facilitated by models can help to develop and optimise therapeutic strategies. Our findings should be considered for the planning of future clinical trials introducing MEKi in the treatment of neuroblastoma.

Dr. Sim: Similarity Learning for Transcriptional Phenotypic Drug discovery

Transcriptional phenotypic drug discovery has achieved great success, and various compound perturbation-based data resources, such as Connectivity Map (CMap) and Library of Integrated Network-Based Cellular Signatures (LINCS), have been presented. Computational strategies fully mining these resources for phenotypic drug discovery have been proposed, and among them, a fundamental issue is to define the proper similarity between the transcriptional profiles to elucidate the drug mechanism of actions and identify new drug indications. Traditionally, this similarity has been defined in an unsupervised way, and due to the high dimensionality and the existence of high noise in those high-throughput data, it lacks robustness with limited performance. In our study, we present Dr. Sim, which is a general learning-based framework that automatically infers similarity measurement rather than being manually designed and can be used to characterize transcriptional phenotypic profiles for drug discovery with generalized good performance. We evaluated Dr. Sim on comprehensively publicly available in vitro and in vivo datasets in drug annotation and repositioning using high-throughput transcriptional perturbation data and indicated that Dr. Sim significantly outperforms the existing methods and is proved to be a conceptual improvement by learning transcriptional similarity to facilitate the broad utility of high-throughput transcriptional perturbation data for phenotypic drug discovery. The source code and usage of Dr. Sim is available at https://github.com/bm2-lab/DrSim/.

Transcriptional signatures of cell-cell interactions are dependent on cellular context

Cell-cell interactions are often predicted from single-cell transcriptomics data based on observing receptor and corresponding ligand transcripts in cells. These predictions could theoretically be improved by inspecting the transcriptome of the receptor cell for evidence of gene expression changes in response to the ligand. It is commonly expected that a given receptor, in response to ligand activation, will have a characteristic downstream gene expression signature. However, this assumption has not been well tested. We used ligand perturbation data from both the high-throughput Connectivity Map resource and published transcriptomic assays of cell lines and purified cell populations to determine whether ligand signals have unique and generalizable transcriptional signatures across biological conditions. Most of the receptors we analyzed did not have such characteristic gene expression signatures - instead these signatures were highly dependent on cell type. Cell context is thus important when considering transcriptomic evidence of ligand signaling, which makes it challenging to build generalizable ligand-receptor interaction signatures to improve cell-cell interaction predictions.

Modeling transcriptional profiles of gene perturbation with deep neural network

Cell line perturbation data could be utilized as a reference for inferring underlying molecular processes in new gene expression profiles. It is important to develop accurate and computationally efficient algorithms to exploit biological knowledge in the growing compendium of existing perturbation data and harness these for new predictions. We reframed the problem of inferring possible gene perturbation based on a reference perturbation database into a classification task and evaluated the application of deep neural network models to address this problem. Our results showed that a fully-connected multi-layer neural network was able to achieve up to 74.9% accuracy in a holdout test set, but the model generalizability was limited by consistency between training and testing data. Capacity and flexibility enables neural network models to efficiently represent transcriptomic features associated with single gene knockdown perturbations. With consistent signals between training and testing sets, neural networks may be trained to classify new samples to experimentally confirmed molecular phenotypes.

Learning epistatic gene interactions from perturbation screens

The treatment of complex diseases often relies on combinatorial therapy, a strategy where drugs are used to target multiple genes simultaneously. Promising candidate genes for combinatorial perturbation often constitute epistatic genes, i.e., genes which contribute to a phenotype in a non-linear fashion. Experimental identification of the full landscape of genetic interactions by perturbing all gene combinations is prohibitive due to the exponential growth of testable hypotheses. Here we present a model for the inference of pairwise epistatic, including synthetic lethal, gene interactions from siRNA-based perturbation screens. The model exploits the combinatorial nature of siRNA-based screens resulting from the high numbers of sequence-dependent off-target effects, where each siRNA apart from its intended target knocks down hundreds of additional genes. We show that conditional and marginal epistasis can be estimated as interaction coefficients of regression models on perturbation data. We compare two methods, namely glinternet and xyz, for selecting non-zero effects in high dimensions as components of the model, and make recommendations for the appropriate use of each. For data simulated from real RNAi screening libraries, we show that glinternet successfully identifies epistatic gene pairs with high accuracy across a wide range of relevant parameters for the signal-to-noise ratio of observed phenotypes, the effect size of epistasis and the number of observations per double knockdown. xyz is also able to identify interactions from lower dimensional data sets (fewer genes), but is less accurate for many dimensions. Higher accuracy of glinternet, however, comes at the cost of longer running time compared to xyz. The general model is widely applicable and allows mining the wealth of publicly available RNAi screening data for the estimation of epistatic interactions between genes. As a proof of concept, we apply the model to search for interactions, and potential targets for treatment, among previously published sets of siRNA perturbation screens on various pathogens. The identified interactions include both known epistatic interactions as well as novel findings.

Neuroblastoma signalling models unveil combination therapies targeting feedback-mediated resistance

Very high risk neuroblastoma is characterised by increased MAPK signalling, and targeting MAPK signalling is a promising therapeutic strategy. We used a deeply characterised panel of neuroblastoma cell lines and found that the sensitivity to MEK inhibitors varied drastically between these cell lines. By generating quantitative perturbation data and mathematical modelling, we determined potential resistance mechanisms. We found that negative feedbacks within MAPK signalling and to the IGF receptor mediate re-activation of MAPK signalling upon treatment in resistant cell lines. By using cell-line specific models, we predict that combinations of MEK inhibitors with RAF or IGFR inhibitors can overcome resistance, and tested these predictions experimentally. In addition, phospo-proteomics profiles confirm the cell-specific feedback effects and synergy of MEK and IGFR targeted treatements. Our study shows that a quantitative understanding of signalling and feedback mechanisms facilitated by models can help to develop and optimise therapeutic strategies, and our findings should be considered for the planning of future clinical trials introducing MEK inhibitors in the treatment of neuroblastoma.

IRF8 Is an AML-Specific Susceptibility Factor That Regulates Signaling Pathways and Proliferation of AML Cells

Personalized treatment of acute myeloid leukemia (AML) that target individual aberrations strongly improved the survival of AML patients. However, AML is still one of the most lethal cancer diseases of the 21st century, demonstrating the need to find novel drug targets and to explore alternative treatment strategies. Upon investigation of public perturbation data, we identified the transcription factor IRF8 as a novel AML-specific susceptibility gene in humans. IRF8 is upregulated in a subset of AML cells and its deletion leads to impaired proliferation in those cells. Consistently, high IRF8 expression is associated with poorer patients’ prognoses. Combining gene expression changes upon IRF8 deletion and the genome-wide localization of IRF8 in the AML cell line MV4-11, we demonstrate that IRF8 directly regulates key signaling molecules, such as the kinases SRC and FAK, the transcription factors RUNX1 and IRF5, and the cell cycle regulator Cyclin D1. IRF8 loss impairs AML-driving signaling pathways, including the WNT, Chemokine, and VEGF signaling pathways. Additionally, many members of the focal adhesion pathway showed reduced expression, providing a putative link between high IRF8 expression and poor prognosis. Thus, this study suggests that IRF8 could serve as a biomarker and potential molecular target in a subset of human AMLs.