Search tips
Search criteria

Results 1-8 (8)

Clipboard (0)

Select a Filter Below

Year of Publication
Document Types
author:("Pal, kanadi")
1.  An Ensemble Based Top Performing Approach for NCI-DREAM Drug Sensitivity Prediction Challenge 
PLoS ONE  2014;9(6):e101183.
We consider the problem of predicting sensitivity of cancer cell lines to new drugs based on supervised learning on genomic profiles. The genetic and epigenetic characterization of a cell line provides observations on various aspects of regulation including DNA copy number variations, gene expression, DNA methylation and protein abundance. To extract relevant information from the various data types, we applied a random forest based approach to generate sensitivity predictions from each type of data and combined the predictions in a linear regression model to generate the final drug sensitivity prediction. Our approach when applied to the NCI-DREAM drug sensitivity prediction challenge was a top performer among 47 teams and produced high accuracy predictions. Our results show that the incorporation of multiple genomic characterizations lowered the mean and variance of the estimated bootstrap prediction error. We also applied our approach to the Cancer Cell Line Encyclopedia database for sensitivity prediction and the ability to extract the top targets of an anti-cancer drug. The results illustrate the effectiveness of our approach in predicting drug sensitivity from heterogeneous genomic datasets.
PMCID: PMC4076307  PMID: 24978814
2.  A Diverse Stochastic Search Algorithm for Combination Therapeutics 
BioMed Research International  2014;2014:873436.
Background. Design of drug combination cocktails to maximize sensitivity for individual patients presents a challenge in terms of minimizing the number of experiments to attain the desired objective. The enormous number of possible drug combinations constrains exhaustive experimentation approaches, and personal variations in genetic diseases restrict the use of prior knowledge in optimization. Results. We present a stochastic search algorithm that consisted of a parallel experimentation phase followed by a combination of focused and diversified sequential search. We evaluated our approach on seven synthetic examples; four of them were evaluated twice with different parameters, and two biological examples of bacterial and lung cancer cell inhibition response to combination drugs. The performance of our approach as compared to recently proposed adaptive reference update approach was superior for all the examples considered, achieving an average of 45% reduction in the number of experimental iterations. Conclusions. As the results illustrate, the proposed diverse stochastic search algorithm can produce optimized combinations in relatively smaller number of iterative steps. This approach can be combined with available knowledge on the genetic makeup of the patient to design optimal selection of drug cocktails.
PMCID: PMC3971504  PMID: 24738075
3.  Integrated Analysis of Transcriptomic and Proteomic Data 
Current Genomics  2013;14(2):91-110.
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.
PMCID: PMC3637682  PMID: 24082820
Integrated omics; Data fusion approaches; Transcriptome; Proteome; Joint modeling; Combined analysis review.
4.  An Adaptive Src-PDGFRA-Raf Axis in Rhabdomyosarcoma 
Alveolar rhabdomyosarcoma (aRMS) is a very aggressive sarcoma of children and young adults. Our previous studies have shown that small molecule inhibition of Pdgfra is initially very effective in an aRMS mouse model. However, slowly evolving, acquired resistance to a narrow-spectrum kinase inhibitor (imatinib) was common. We identified Src family kinases (SFKs) to be potentiators of Pdgfra in murine aRMS primary cell cultures from mouse tumors with evolved resistance in vivo in comparison to untreated cultures. Treating the resistant primary cell cultures with a combination of Pdgfra and Src inhibitors had a strong additive effect on cell viability. In Pdgfra knockout tumors, however, the Src inhibitor had no effect on tumor cell viability. Sorafenib, whose targets include not only PDGFRA but also the Src downstream target Raf, was effective at inhibiting mouse and human tumor cell growth and halted progression of mouse aRMS tumors in vivo. These results suggest that an adaptive Src-Pdgfra-Raf-Mapk axis is relevant to PDGFRA inhibition in rhabdomyosarcoma.
PMCID: PMC3463776  PMID: 22960170
Imatinib; Sorafenib; receptor tyrosine kinase; Pdgfra
5.  A new approach for prediction of tumor sensitivity to targeted drugs based on functional data 
BMC Bioinformatics  2013;14:239.
The success of targeted anti-cancer drugs are frequently hindered by the lack of knowledge of the individual pathway of the patient and the extreme data requirements on the estimation of the personalized genetic network of the patient’s tumor. The prediction of tumor sensitivity to targeted drugs remains a major challenge in the design of optimal therapeutic strategies. The current sensitivity prediction approaches are primarily based on genetic characterizations of the tumor sample. We propose a novel sensitivity prediction approach based on functional perturbation data that incorporates the drug protein interaction information and sensitivities to a training set of drugs with known targets.
We illustrate the high prediction accuracy of our framework on synthetic data generated from the Kyoto Encyclopedia of Genes and Genomes (KEGG) and an experimental dataset of four canine osteosarcoma tumor cultures following application of 60 targeted small-molecule drugs. We achieve a low leave one out cross validation error of <10% for the canine osteosarcoma tumor cultures using a drug screen consisting of 60 targeted drugs.
The proposed framework provides a unique input-output based methodology to model a cancer pathway and predict the effectiveness of targeted anti-cancer drugs. This framework can be developed as a viable approach for personalized cancer therapy.
PMCID: PMC3750584  PMID: 23890326
6.  Boolean network inference from time series data incorporating prior biological knowledge 
BMC Genomics  2012;13(Suppl 6):S9.
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. 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.
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. 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.
PMCID: PMC3481452  PMID: 23134816
8.  Evidence for an Unanticipated Relationship Between Undifferentiated Pleomorphic Sarcoma and Embryonal Rhabdomyosarcoma 
Cancer cell  2011;19(2):177-191.
Embryonal rhabdomyosarcoma (eRMS) shows the most myodifferentiation amongst sarcomas, yet the precise cell of origin remains undefined. Using Ptch1, p53 and/or Rb1 conditional mouse models and controlling prenatal or postnatal myogenic cell of origin, we demonstrate that eRMS and undifferentiated pleomorphic sarcoma (UPS) lie in a continuum, with satellite cells predisposed to giving rise to UPS. Conversely, p53 loss in maturing myoblasts gives rise to eRMS, which have the highest myodifferentiation potential. Irrespective of origin, Rb1 loss modifies tumor phenotype to mimic UPS. In human sarcomas that lack pathognomic chromosomal translocations, p53 loss of function is prevalent whereas Shh or Rb1 alterations likely act primarily as modifiers. Thus, sarcoma phenotype is strongly influenced by cell of origin and mutational profile.
PMCID: PMC3040414  PMID: 21316601

Results 1-8 (8)