Signaling network prediction by the Ontology Fingerprint enhanced Bayesian network

Bayesian network enables interpretable and state-of-the-art prediction of immunotherapy responses in cancer patients

Immune checkpoint inhibitors, especially PD-1/PD-L1 blockade, have revolutionized cancer treatment and brought tremendous benefits to patients who otherwise would have had a limited prognosis. Nonetheless, only a small fraction of patients respond to immunotherapy, and the costs and side effects of immune checkpoint inhibitors cannot be ignored. With the advent of machine and deep learning, clinical and genetic data have been used to stratify patient responses to immunotherapy. Unfortunately, these approaches have typically been "black-box" methods that are unable to explain their predictions, thereby hindering their responsible clinical application. Herein, we developed a "white-box" Bayesian network model that achieves accurate and interpretable predictions of immunotherapy responses against nonsmall cell lung cancer (NSCLC). This tree-augmented naïve Bayes (TAN) model accurately predicted durable clinical benefits and distinguished two clinically significant subgroups with distinct prognoses. Furthermore, our state-of-the-art white-box TAN approach achieved greater accuracy than previous methods. We hope that our model will guide clinicians in selecting NSCLC patients who truly require immunotherapy and expect our approach to be easily applied to other types of cancer.

Evidence for craniofacial enhancer variation underlying nonsyndromic cleft lip and palate

Nonsyndromic cleft lip with or without cleft palate (NSCLP) is a common birth defect for which only ~ 20% of the underlying genetic variation has been identified. Variants in noncoding regions have been increasingly suggested to contribute to the missing heritability. In this study, we investigated whether variation in craniofacial enhancers contributes to NSCLP. Candidate enhancers were identified using VISTA Enhancer Browser and previous publications. Prioritization was based on patterning defects in knockout mice, deletion/duplication of craniofacial genes in animal models and results of whole exome/whole genome sequencing studies. This resulted in 20 craniofacial enhancers to be investigated. Custom amplicon-based sequencing probes were designed and used for sequencing 380 NSCLP probands (from multiplex and simplex families of non-Hispanic white (NHW) and Hispanic ethnicities) using Illumina MiSeq. The frequencies of identified variants were compared to ethnically matched European (CEU) and Los Angeles Mexican (MXL) control genomes and used for association analyses. Variants in mm427/MSX1 and hs1582/SPRY1 showed genome-wide significant association with NSCLP (p ≤ 6.4 × 10^-11). In silico analysis showed that these enhancer variants may disrupt important transcription factor binding sites. Haplotypes involving these enhancers and also mm435/ABCA4 were significantly associated with NSCLP, especially in NHW (p ≤ 6.3 × 10^-7). Importantly, groupwise burden analysis showed several enhancer combinations significantly over-represented in NSCLP individuals, revealing novel NSCLP pathways and supporting a polygenic inheritance model. Our findings support the role of craniofacial enhancer sequence variation in the etiology of NSCLP.

Deep learning for drug response prediction in cancer

Predicting the sensitivity of tumors to specific anti-cancer treatments is a challenge of paramount importance for precision medicine. Machine learning(ML) algorithms can be trained on high-throughput screening data to develop models that are able to predict the response of cancer cell lines and patients to novel drugs or drug combinations. Deep learning (DL) refers to a distinct class of ML algorithms that have achieved top-level performance in a variety of fields, including drug discovery. These types of models have unique characteristics that may make them more suitable for the complex task of modeling drug response based on both biological and chemical data, but the application of DL to drug response prediction has been unexplored until very recently. The few studies that have been published have shown promising results, and the use of DL for drug response prediction is beginning to attract greater interest from researchers in the field. In this article, we critically review recently published studies that have employed DL methods to predict drug response in cancer cell lines. We also provide a brief description of DL and the main types of architectures that have been used in these studies. Additionally, we present a selection of publicly available drug screening data resources that can be used to develop drug response prediction models. Finally, we also address the limitations of these approaches and provide a discussion on possible paths for further improvement. Contact:  mrocha@di.uminho.pt.

MGT-SM: A Method for Constructing Cellular Signal Transduction Networks

A cellular signal transduction network is an important means to describe biological responses to environmental stimuli and exchange of biological signals. Constructing the cellular signal transduction network provides an important basis for the study of the biological activities, the mechanism of the diseases, drug targets and so on. The statistical approaches to network inference are popular in literature. Granger test has been used as an effective method for causality inference. Compared with bivariate granger tests, multivariate granger tests reduce the indirect causality and were used widely for the construction of cellular signal transduction networks. A multivariate Granger test requires that the number of time points in the time-series data is more than the number of nodes involved in the network. However, there are many real datasets with a few time points which are much less than the number of nodes in the network. In this study, we propose a new multivariate Granger test-based framework to construct cellular signal transduction network, called MGT-SM. Our MGT-SM uses SVD to compute the coefficient matrix from gene expression data and adopts Monte Carlo simulation to estimate the significance of directed edges in the constructed networks. We apply the proposed MGT-SM to Yeast Synthetic Network and MDA-MB-468, and evaluate its performance in terms of the recall and the AUC. The results show that MGT-SM achieves better results, compared with other popular methods (CGC2SPR, PGC, and DBN).

Gene fingerprint model for literature based detection of the associations among complex diseases: a case study of COPD

BACKGROUND

Disease comorbidity is very common and has significant impact on disease treatment. Revealing the associations among diseases may help to understand the mechanisms of diseases, improve the prevention and treatment of diseases, and support the discovery of new drugs or new uses of existing drugs.

METHODS

In this paper, we introduced a mathematical model to represent gene related diseases with a series of associated genes based on the overrepresentation of genes and diseases in PubMed literature. We also illustrated an efficient way to reveal the implicit connections between COPD and other diseases based on this model.

RESULTS

We applied this approach to analyze the relationships between Chronic Obstructive Pulmonary Disease (COPD) and other diseases under the Lung diseases branch in the Medical subject heading index system and detected 4 novel diseases relevant to COPD. As judged by domain experts, the F score of our approach is up to 77.6%.

CONCLUSIONS

The results demonstrate the effectiveness of the gene fingerprint model for diseases on the basis of medical literature.

A Multi-Parameter Analysis of Cellular Coordination of Major Transcriptome Regulation Mechanisms

To understand cellular coordination of multiple transcriptome regulation mechanisms, we simultaneously measured transcription rate (TR), mRNA abundance (RA) and translation activity (TA). This revealed multiple insights. First, the three parameters displayed systematic statistical differences. Sequentially more genes exhibited extreme (low or high) expression values from TR to RA, and then to TA; that is, cellular coordination of multiple transcriptome regulatory mechanisms leads to sequentially enhanced gene expression selectivity as the genetic information flow from the genome to the proteome. Second, contribution of the stabilization-by-translation regulatory mechanism to the cellular coordination process was assessed. The data enabled an estimation of mRNA stability, revealing a moderate but significant positive correlation between mRNA stability and translation activity. Third, the proportion of mRNA occupied by un-translated regions (UTR) exhibited a negative relationship with the level of this correlation, and was thus a major determinant of the mode of regulation of the mRNA. High-UTR-proportion mRNAs tend to defy the stabilization-by-translation regulatory mechanism, staying out of the polysome but remaining stable; mRNAs with little UTRs largely followed this regulation. In summary, we quantitatively delineated the relationship among multiple transcriptome regulation parameters, i.e., cellular coordination of corresponding regulatory mechanisms.

Integrating network reconstruction with mechanistic modeling to predict cancer therapies

Signal transduction networks are often rewired in cancer cells. Identifying these alterations will enable more effective cancer treatment. We developed a computational framework that can identify, reconstruct, and mechanistically model these rewired networks from noisy and incomplete perturbation response data and then predict potential targets for intervention. As a proof of principle, we analyzed a perturbation data set targeting epidermal growth factor receptor (EGFR) and insulin-like growth factor 1 receptor (IGF1R) pathways in a panel of colorectal cancer cells. Our computational approach predicted cell line-specific network rewiring. In particular, feedback inhibition of insulin receptor substrate 1 (IRS1) by the kinase p70S6K was predicted to confer resistance to EGFR inhibition, suggesting that disrupting this feedback may restore sensitivity to EGFR inhibitors in colorectal cancer cells. We experimentally validated this prediction with colorectal cancer cell lines in culture and in a zebrafish (Danio rerio) xenograft model.

Study of phosphorylation events for cancer diagnoses and treatment

The activation of signaling cascades in response to extracellular and intracellular stimuli to control cell growth, proliferation and survival, is orchestrated by protein kinases via phosphorylation. A critical issue is the study of the mechanisms of cancer cells for the development of more effective drugs. With the application of the new proteomic technologies, together with the advancement in the sequencing of the human proteome, patients will therefore be benefited by the discovery of novel therapeutic and/or diagnostic protein targets. Furthermore, the advances in proteomic approaches and the Human Proteome Organization (HUPO) have opened a new door which is helpful in the identification of patients at risk and towards improving current therapies. Modification of the signaling-networks via mutations or abnormal protein expression underlies the cause or consequence of many diseases including cancer. Resulting data is used to reveal connections between genes proteins and compounds and the related molecular pathways for underlining disease states. As a delegate of HUPO, for human proteome on children assays and studies, we, at Hospital Universitario Niño Jesús, are seeking to support the human proteome in this context. Clinical goals have to be clearly established and proteomics experts have to set up the appropriate proteomic strategy, which coupled to bioinformatics will make it possible to achieve new therapies for patients with poor prognosis. We envision to combine our up-coming data to the HUPO organization in order to support international efforts to advance the cure of cancer disease.

An Integrated Qualitative and Quantitative Biochemical Model Learning Framework Using Evolutionary Strategy and Simulated Annealing

Both qualitative and quantitative model learning frameworks for biochemical systems have been studied in computational systems biology. In this research, after introducing two forms of pre-defined component patterns to represent biochemical models, we propose an integrative qualitative and quantitative modelling framework for inferring biochemical systems. In the proposed framework, interactions between reactants in the candidate models for a target biochemical system are evolved and eventually identified by the application of a qualitative model learning approach with an evolution strategy. Kinetic rates of the models generated from qualitative model learning are then further optimised by employing a quantitative approach with simulated annealing. Experimental results indicate that our proposed integrative framework is feasible to learn the relationships between biochemical reactants qualitatively and to make the model replicate the behaviours of the target system by optimising the kinetic rates quantitatively. Moreover, potential reactants of a target biochemical system can be discovered by hypothesising complex reactants in the synthetic models. Based on the biochemical models learned from the proposed framework, biologists can further perform experimental study in wet laboratory. In this way, natural biochemical systems can be better understood.

Using Ontology Fingerprints to disambiguate gene name entities in the biomedical literature

Ambiguous gene names in the biomedical literature are a barrier to accurate information extraction. To overcome this hurdle, we generated Ontology Fingerprints for selected genes that are relevant for personalized cancer therapy. These Ontology Fingerprints were used to evaluate the association between genes and biomedical literature to disambiguate gene names. We obtained 93.6% precision for the test gene set and 80.4% for the area under a receiver-operating characteristics curve for gene and article association. The core algorithm was implemented using a graphics processing unit-based MapReduce framework to handle big data and to improve performance. We conclude that Ontology Fingerprints can help disambiguate gene names mentioned in text and analyse the association between genes and articles.

Database URL: http://www.ontologyfingerprint.org

Modeling signal transduction from protein phosphorylation to gene expression

BACKGROUND

Signaling networks are of great importance for us to understand the cell’s regulatory mechanism. The rise of large-scale genomic and proteomic data, and prior biological knowledge has paved the way for the reconstruction and discovery of novel signaling pathways in a data-driven manner. In this study, we investigate computational methods that integrate proteomics and transcriptomic data to identify signaling pathways transmitting signals in response to specific stimuli. Such methods can be applied to cancer genomic data to infer perturbed signaling pathways.

METHOD

We proposed a novel Bayesian Network (BN) framework to integrate transcriptomic data with proteomic data reflecting protein phosphorylation states for the purpose of identifying the pathways transmitting the signal of diverse stimuli in rat and human cells. We represented the proteins and genes as nodes in a BN in which edges reflect the regulatory relationship between signaling proteins. We designed an efficient inference algorithm that incorporated the prior knowledge of pathways and searched for a network structure in a data-driven manner.

RESULTS

We applied our method to infer rat and human specific networks given gene expression and proteomic datasets. We were able to effectively identify sparse signaling networks that modeled the observed transcriptomic and proteomic data. Our methods were able to identify distinct signaling pathways for rat and human cells in a data-driven manner, based on the facts that rat and human cells exhibited distinct transcriptomic and proteomics responses to a common set of stimuli. Our model performed well in the SBV IMPROVER challenge in comparison to other models addressing the same task. The capability of inferring signaling pathways in a data-driven fashion may contribute to cancer research by identifying distinct aberrations in signaling pathways underlying heterogeneous cancers subtypes.

Computational methods and opportunities for phosphorylation network medicine

Protein phosphorylation, one of the most ubiquitous post-translational modifications (PTM) of proteins, is known to play an essential role in cell signaling and regulation. With the increasing understanding of the complexity and redundancy of cell signaling, there is a growing recognition that targeting the entire network or system could be a necessary and advantageous strategy for treating cancer. Protein kinases, the proteins that add a phosphate group to the substrate proteins during phosphorylation events, have become one of the largest groups of 'druggable' targets in cancer therapeutics in recent years. Kinase inhibitors are being regularly used in clinics for cancer treatment. This therapeutic paradigm shift in cancer research is partly due to the generation and availability of high-dimensional proteomics data. Generation of this data, in turn, is enabled by increased use of mass-spectrometry (MS)-based or other high-throughput proteomics platforms as well as companion public databases and computational tools. This review briefly summarizes the current state and progress on phosphoproteomics identification, quantification, and platform related characteristics. We review existing database resources, computational tools, methods for phosphorylation network inference, and ultimately demonstrate the connection to therapeutics. Finally, many research opportunities exist for bioinformaticians or biostatisticians based on developments and limitations of the current and emerging technologies.

Advances in systems biology: computational algorithms and applications

The 2012 International Conference on Intelligent Biology and Medicine (ICIBM 2012) was held on April 22-24, 2012 in Nashville, Tennessee, USA. The conference featured six technical sessions, one tutorial session, one workshop, and 3 keynote presentations that covered state-of-the-art research activities in genomics, systems biology, and intelligent computing. In addition to a major emphasis on the next generation sequencing (NGS)-driven informatics, ICIBM 2012 aligned significant interests in systems biology and its applications in medicine. We highlight in this editorial the selected papers from the meeting that address the developments of novel algorithms and applications in systems biology.