Oncological pathologies are brought forth by multiple confederating genetic, proteomic and cellular disorders. To understand the onset and progression of such oncological manifestations, the underlying participatory biological players need to be integratively investigated. To that affect, only computational tools can integrate such spatiotemporal cancer biology data, thus leading to the creation of unified and coherent models for further analysis.
The salient modelling tools developed so far for this purpose include CHASTE, CompuCell, ELECANS, and Repast etc. Although, these platforms offer varying degrees of modelling coverage, and next-generation oncology data integration, however, their resulting models lack conformation and standardization. Furthermore, the modelling strategies and the numerical algorithms used and the results thus obtained from these software cannot be benchmarked for reproduction and verification. Also, the front-end user-interfaces offered by these platforms are unintuitive and not conducive to use by a layman researcher. Last but not the least, the simulation runtime performances of these platforms are not at pace for modelling biological entities in real time. Besides these major shortcomings, there are numerous other associated challenges (e.g. platform scalability, modelling flexibility, and execution of custom user codes) that need to be overcome before a realistically detailed multiscale cancer model can be constructed and simulated in real-time.
Towards addressing these issues, during the course of this project, we would like to develop a next-generation multiscale cancer modeling platform. The envisaged platform will not only improve upon the pre-existing modelling features present in the aforementioned platforms (along with overcoming their shortcomings), but also introduce several novel modelling and simulation strategies. The novel features proposed for implementation in a next-generation modelling platform include (i) an integrated gene and protein network editor, (ii) a cell cycle designer, (iii) a tissue designer and assembler, (iv) a GPU-based high performance core, (v) a comprehensive ‘What You See Is What You Get’ (WYSIWYG) graphical user interface accompanied with a software development kit, and (vi) a wet-lab data integration framework.
Upon its completion, the proposed multiscale cancer modelling platform will be the state of the art software in cancer modelling which will stand to deliver major advantages to the cancer patients, researchers and the pharmaceutical industry. The patients will be benefited by the personalized medicines and therapeutics developed by modelling and analysis of their pathological data using the proposed platform. The cancer researchers stand to gain by obtaining a computational modelling framework using which they can save precious wet-lab materials and resources. The pharmaceutical industry can use this platform to investigate novel drug targets from personalized patient data and design newer drugs for treatment of cancer. Additionally, the students at LUMS and other partnering institutes will have perpetual access to a state of the art multiscale cancer modelling software and associated code to experiment and obtain rich experience during their studies.
This project encompasses the design and development of a next generation multiscale modelling platform for applications in cancer systems biology. This modelling platform envisages a seamless integration of next generation sequencing and quantitation data from the wet labs and its onward usage for investigating the roles of known oncological factors. A powerful HPC enabled simulation engine will be developed and coupled with an intuitive and customizable GUI. The resulting software and case study model will be made available to the laboratories, research centers, universities as well as the pharmaceutical companies which are aiming to develop or evaluate therapeutic targets as well as drugs for treatment of cancer patients. Upon deployment, the proposed modelling platform will not only assist in the modelling of next generation cancer systems biology data, but will also significantly enhance the throughput of the model building and simulation processes. Along side, we are also building case study models of Colorectal Cancer, Pancreatic Cancer and the Warburg Effect for deployment using the aforementioned modelling pipeline.
We are also developing a SPECTRUM - a MATLAB toolbox and PERCPTRON - a web-based protein sequence search engine that can optimally leverage high resolution mass spectral data and act as a platform to seed and stir computational top-down (whole protein) and bottom-up (peptides) proteomics research in Pakistan. The resulting search engine will be available, free of cost, to the experimental and in silico biologists across the country. Moreover, this project will also act as a collaboration avenue between academia and industry, wherein industrial partners can also leverage the software to analyse their data and provide valuable feedback towards developing intelligent algorithms geared towards an improved biomolecular identification and characterisation. To demonstrate the utlity of these search engines, we are undertaking metaproteomics case studies of the Hadiara Drain in Lahore, Pakistan.
Hepatitis C Virus (HCV) infects between 15 to 30% of Pakistani population. At BIRL we are aiming to design new theraputic strategies which employ a direct acting strategy to clear the virus. We employ a mix of wet-lab and dry-lab techniques to identify targetable regions on the virus and test them in vitro.
Another ongoing project at BIRL is the systematic analysis of pregnancy and associated symptoms from patients across Lahore towards devising antenatal e-healthcare for rural areas of Pakistan.The provision of Antenatal Care (ANC) for pregnant women plays a vital role in ensuring infant and maternal health. Limited access to antenatal care in Low and Middle Income Countries (LMIC) results in high Infant and Maternal Mortality Rate (IMR and MMR, respectively). In this project, we are developing a cloud-based clinical Decision Support System (DSS) integrated with a wearable health-sensor network for patient self-diagnosis and real time health monitoring. Patient assessment will be performed by evaluating the human-input coupled with sensor-generated symptomatic information using a Bayesian network driven DSS. High risk pregnancies can be identified and monitored along with dispensing of consultant advice directly to the patient. Patient and disease incidence data is stored on the cloud for tuning probabilities of the Bayesian network towards improving accuracy of predicting anomalies within the epidemiological context. The system therefore, aims to control IMR and MMR by providing ubiquitous access to ANC in LMICs. A scaled-up implementation of the proposed system can help reduce patient influx at the limited tertiary care centers by referring low-risk cases to primary or secondary care establishments.