Saad Haider

Integrated Analysis of Transcriptomic and Proteomic Data

Until recently, understanding the regulatory behavior of cells has been pursued through independent analysis of the transcriptome or the proteome. Based on the central dogma, it was generally assumed that there exist a direct correspondence between mRNA transcripts and generated protein expressions. However, recent studies have shown that the correlation between mRNA and Protein expressions can be low due to various factors such as different half lives and post transcription machinery. Thus, a joint analysis of the transcriptomic and proteomic data can provide useful insights that may not be deciphered from individual analysis of mRNA or protein expressions. This article reviews the existing major approaches for joint analysis of transcriptomic and proteomic data. We categorize the different approaches into eight main categories based on the initial algorithm and final analysis goal. We further present analogies with other domains and discuss the existing research problems in this area.

A Copula Based Approach for Design of Multivariate Random Forests for Drug Sensitivity Prediction

Modeling sensitivity to drugs based on genetic characterizations is a significant challenge in the area of systems medicine. Ensemble based approaches such as Random Forests have been shown to perform well in both individual sensitivity prediction studies and team science based prediction challenges. However, Random Forests generate a deterministic predictive model for each drug based on the genetic characterization of the cell lines and ignores the relationship between different drug sensitivities during model generation. This application motivates the need for generation of multivariate ensemble learning techniques that can increase prediction accuracy and improve variable importance ranking by incorporating the relationships between different output responses. In this article, we propose a novel cost criterion that captures the dissimilarity in the output response structure between the training data and node samples as the difference in the two empirical copulas. We illustrate that copulas are suitable for capturing the multivariate structure of output responses independent of the marginal distributions and the copula based multivariate random forest framework can provide higher accuracy prediction and improved variable selection. The proposed framework has been validated on genomics of drug sensitivity for cancer and cancer cell line encyclopedia database.

An integrated approach to anti-cancer drug sensitivity prediction

A framework for design of personalized cancer therapy requires the ability to predict the sensitivity of a tumor to anticancer drugs. The predictive modeling of tumor sensitivity to anti-cancer drugs has primarily focused on generating functions that map gene expressions and genetic mutation profiles to drug sensitivity. In this paper, we present a new approach for drug sensitivity prediction and combination therapy design based on integrated functional and genomic characterizations. The modeling approach when applied to data from the Cancer Cell Line Encyclopedia shows a significant gain in prediction accuracy as compared to elastic net and random forest techniques based on genomic characterizations. Utilizing a Mouse Embryonal Rhabdomyosarcoma cell culture and a drug screen of 60 targeted drugs, we show that predictive modeling based on functional data alone can also produce high accuracy predictions. The framework also allows us to generate personalized tumor proliferation circuits to gain further insights on the individualized biological pathway.

Boolean network inference from time series data incorporating prior biological knowledge

BackgroundNumerous approaches exist for modeling of genetic regulatory networks (GRNs) but the low sampling rates often employed in biological studies prevents the inference of detailed models from experimental data. In this paper, we analyze the issues involved in estimating a model of a GRN from single cell line time series data with limited time points.ResultsWe present an inference approach for a Boolean Network (BN) model of a GRN from limited transcriptomic or proteomic time series data based on prior biological knowledge of connectivity, constraints on attractor structure and robust design. We applied our inference approach to 6 time point transcriptomic data on Human Mammary Epithelial Cell line (HMEC) after application of Epidermal Growth Factor (EGF) and generated a BN with a plausible biological structure satisfying the data. We further defined and applied a similarity measure to compare synthetic BNs and BNs generated through the proposed approach constructed from transitions of various paths of the synthetic BNs. We have also compared the performance of our algorithm with two existing BN inference algorithms.ConclusionsThrough theoretical analysis and simulations, we showed the rarity of arriving at a BN from limited time series data with plausible biological structure using random connectivity and absence of structure in data. The framework when applied to experimental data and data generated from synthetic BNs were able to estimate BNs with high similarity scores. Comparison with existing BN inference algorithms showed the better performance of our proposed algorithm for limited time series data. The proposed framework can also be applied to optimize the connectivity of a GRN from experimental data when the prior biological knowledge on regulators is limited or not unique.

Inference of tumor inhibition pathways from drug perturbation data

The tumor proliferation pathways for each individual patient encompass variations and a successful treatment regime based on targeted drugs necessitates the estimation of the influences of target inhibition on cell viability. In this article, we consider an inference approach to decipher the significant blocks of protein targets and the effect of their inhibition on tumor proliferation. Our framework is based on sequential search and non-linear optimization for estimating the block parameters. The proposed algorithm is tested on extensive synthetic data and provides high accuracy estimates for model parameters. We furthermore evaluated the performance of the framework in presence of noise and were able to achieve high precision cell viability prediction.

Anticancer drug sensitivity analysis: An integrated approach applied to Erlotinib sensitivity prediction in the CCLE database

ABSTRACT The cancer cell line encyclopedia (CCLE), a joint academic and industry collaboration, provides a vast resource for analyzing the effectiveness of anti-cancer drugs across numerous cell lines. The predictive modeling of tumor sensitivity to targeted drugs has primarily focused on generating functions that map gene expressions and genetic mutation profiles to targeted drug sensitivity. The prediction accuracies of genomic signature based models are often limited as reported in initial analysis (Barretina et al.) of CCLE database. In this article, we illustrate that incorporating the target inhibition profile of the anticancer drugs and the functional behavior of related drugs will enable us to achieve much higher prediction accuracy.

Combination therapy design for targeted therapeutics from a drug-protein interaction perspective

ABSTRACT In the last decade, a number of drugs targeting specific proteins have been developed that are becoming common in cancer research as a basis for personalized therapy. How-ever, the numerous aberrations in molecular pathways that can produce cancer necessitate the use of drug combinations as compared to single drugs for treatment of individual cancers. In this article, we consider the design of combination therapy based on tumor sensitivity measurements over a panel of targeted drugs. We consider the following two optimization criteria (a) generating drug combinations with high sensitivity and minimal toxicity and (b) generating drug combinations targeting multiple parallel pathways for avoiding resistance. The optimization problem is solved using a set cover approach and a sequential search hill climbing technique. The effectiveness of our optimization procedure is illustrated on both synthetic and experimental models.

Inference of a genetic regulatory network model from limited time series data

Numerous approaches exist for modeling of genetic regulatory networks (GRNs) but the low sampling rates often employed in biological studies prevents the inference of detailed models from experimental data. In this paper, we analyze the issues involved in estimating a model of a GRN from single cell line time series data with limited time points. We present an inference approach for a Boolean Network (BN) model of a GRN from limited transcriptomic or proteomic time series data based on prior biological knowledge of connectivity, constraints on attractor structure and robust design. Through theoretical analysis and simulations, we showed the rarity of arriving at a BN from limited time series data with plausible biological structure using random connectivity and absence of structure in data. We applied our inference approach to 6 time point transcriptomic data on HMEC cell lines after application of EGF and generated a BN with a plausible biological structure satisfying the data.

Design of Probabilistic Random Forests with Applications to Anticancer Drug Sensitivity Prediction

Random forests consisting of an ensemble of regression trees with equal weights are frequently used for design of predictive models. In this article, we consider an extension of the methodology by representing the regression trees in the form of probabilistic trees and analyzing the nature of heteroscedasticity. The probabilistic tree representation allows for analytical computation of confidence intervals (CIs), and the tree weight optimization is expected to provide stricter CIs with comparable performance in mean error. We approached the ensemble of probabilistic trees’ prediction from the perspectives of a mixture distribution and as a weighted sum of correlated random variables. We applied our methodology to the drug sensitivity prediction problem on synthetic and cancer cell line encyclopedia dataset and illustrated that tree weights can be selected to reduce the average length of the CI without increase in mean error.

Quantifying the inference power of a drug screen for predictive analysis

A model for drug sensitivity prediction is often inferred from the response of a training drug screen. Quantifying the inference power of perturbations before experimentation will assist in selecting drugs screens with higher predictive power. In this article, we present a novel approach to quantify the inference power of a drug screen based on drug target profiles and biologically motivated monotonicity constraints. We have tested our algorithm on synthetically and experimentally generated datasets and the results illustrate the suitability of the proposed measure in estimating information gained from an experimental drug screen.