Pancreatic cancer is extremely difficult to treat with only about 5 % of patients surviving for 5 years. New treatments are urgently needed and many cancer researchers are developing and testing new therapies. These researchers often use specially bred mice to study pancreatic cancer. These mice are born with a healthy pancreas but develop pancreatic tumours deep within their body as they age. Its possible to detect these pancreatic tumours from outside the body using imaging and this can be done so it has very little effect on the animal. Imaging allows scientists to study the tumour growth and whether or not it responds to new therapies. However, pancreatic tumours can be extremely hard to detect for scientists who are not experts in imaging. Therefore we have developed an artificial intelligence computer program that can find pancreatic tumours on MRI scans automatically with very little input from the researcher. This allows for faster and more accurate tumour detection. Because the tumour can be measured more accurately we can reduce the numbers of animals used in the study and detect tumours earlier when they have less impact on the animal's health. The small MRI instrument that was used to develop the computer program allows scientist to perform the scans themselves after minimal training. In this project, we are transferring the method to accurately measure the tumours to three other cancer institutes in the UK. We think that using the artificial intelligence program will allow them also to improve the accuracy of their studies while also reducing the numbers of animals they use in pancreatic cancer research.

The synergistic advances that can be made by the multidisciplinary interplay between abstracted computational modelling and biological experimental investigation within a system biology context are poised to make major contributions to our understanding of some of the most important biological systems implicated in the genesis of many serious diseases such as cancer. However, due to the unavoidable inherent levels of uncertainty, noise and relative scarcity of biological data it is vital that sound evidential based scientific reasoning be enabled within a systems biology context by formally embedding mechanistic models within a probabilistic inferential framework. The synthesis of mechanistic modelling & probabilistic inference provides outstanding opportunities to make further significant advances in understanding biological systems and processes at multiple levels, by defining system components and inferring how they dynamically interact. There is a major role that statistical machine learning methodology has to play in both computational & systems biology research and a number of important methodological challenges are presented by applications working at this interface.However, one of the most important aspects of successful computational & systems biology research is that it must be conducted in direct collaboration with world-class experimental biologists. An outstanding feature of this Fellowship is that it has set in place six exciting collaborations with internationally leading cancer researchers, proteomics technologists, biochemists and plant biologists who are all fully committed to successfully driving forward a potentially groundbreaking multidisciplinary systems biology research programme as detailed in this proposal. Three important application areas within biological science will shape and direct the research to be undertaken during this Fellowship. The applications are distinct, yet overlap in terms of the modelling & inferential issues which each present and this is important in ensuring a consistent and coherent line of research. They have also been selected for their major importance in the study of cellular mechanisms which are fundamental to cell function, some of which are implicated in certain serious diseases. In addition, the applicant has substantive ongoing collaborations with world-class laboratories engaged in these biological investigations. This ensures the proposed research programme is focused on realistic methodological problems which will have a direct impact on the major scientific questions being asked within each area, as well contributing to the computational and inferential sciences. The first application will develop the inferential tools required by cancer biologists when reasoning about the structures underlying the observed dynamics of the MAPK pathway and these tools will be employed in a large scale study of this pathway in collaboration with the Beatson Institute of Cancer Research. The second application, to be conducted with the Plant Sciences group at the University of Glasgow, will seek to elucidate, in a model-based inferential manner, the remarkable observed phenomenon of organ specificity of the circadian clock in soybean and Arabidopsis, in addition a study of models of transcriptional regulation in the cell-cycle will be conducted. The final application will investigate a number of open issues associated with clinical transcriptomics and proteomics where the identification of possible target genes and proteins is of vital importance to cancer researchers in their studies of, in this case breast and ovarian cancer. This study will be conducted in direct conjunction with the Institute of Cancer Research where an ongoing study of BRCA1&2 mutations implicated in breast and ovarian cancer is underway.

The tremendous technological advance underlying the ongoing genomic revolution in the life sciences has profoundly transformed biological research over the last 10-15 years. The grand challenge ahead is to leverage experimental progress like high-throughput sequencing by developing effective tools for large scale data-based inference and optimization. The goals of INFERNET rely on the transfer of ideas and methods developed in recent years at the boundary between statistical physics and information theory into the world of quantitative biology. We aim at setting up a consortium characterized by a proven track-record of high-quality research in order to implement a highly integrated program leading from the design of new algorithms to concrete biological applications. The consortium will provide the optimal environment to nurture a generation of researchers that will drive new developments at the forefront of these challenging fields. The perimeter of each individual research activity will be delimited by: (a) the research themes characterized by the toolbox and methods developed and shared within INFERNET, (b) the choice of the application domains. Principal research themes covered by the consortium will be: (i) the inference of interaction networks from data, (ii) the analysis of static and dynamical processes on networks. Application domains can be broken down into four main areas: (i) the inference and modeling of multi-scale biological networks, (ii) the rational design of biological molecules, (iii) the quantitative study of cell energetics in proliferative regimes, (iv) the characterization of functional states of large-scale regulatory networks. Each individual research project will be a puzzle piece of the wider research project, as well as tailored to suit each researcher's scientific and professional development.

Approximately ninety percent of all cancer-related deaths are caused by the spread of cancer cells to distant sites rather than the growth of the primary tumour. The process of metastasis formation requires the ability of tumour cells to spread from primary tumours, by invasion of surrounding tissue, and afterwards by spreading to distant organs in order to manifest its devastating consequences for patient survival. The series of distinct steps leading to the formation of metastases has been referred to as the “invasion-metastasis-cascade”. For many years there has been an absence of genetically engineered mouse models of invasive intestinal cancer to analyse the complex processes occurring during the invasion-metastasis-cascade and the presumably underlying mechanisms such as epithelial-mesenchymal transition (EMT), cancer stem cell populations and mutation rates in the intestine. Combining adenomatous polyposis coli (Apc)-loss with either p53 mutation or Tgf-β receptor (Tgfbr)-loss results in invasive intestinal carcinoma of mice. The host laboratory generated these two contrasting models of invasive adenocarcinoma, which will be utilized to generate an invasion signature of mRNA and miRNA expression to analyse the underlying processes and important players and targets for future therapies of this initial step of the invasion-metastasis-cascade. Additionally, by comparing the obtained expression profiles to previously published intestinal stem cell signatures and by using genetic labelling in carcinomas, I will analyse the cell culture and human tissue based theories, which indicate that invasive tumour cells, which underwent an EMT, display increased stem cell properties and numbers. This knowledge will generate important information for future cancer stem cell targeting therapeutic strategies.