Pakistan's economy is ranked 26th in the world in terms of purchasing power parity (PPP) and 40th in terms of nominal gross domestic product. Pakistan has a population of approximately 190 million people, making it the world's sixth-largest country, with a nominal GDP per capita of $1,427, ranking 133rd globally. The key sectors of the Pakistani economy are agriculture, mining, industry, automotive, construction, defence, services, and transportation. The IT/ITeS sector is one of Pakistan's fastest-growing industries, accounting for around 1% of the country's GDP ($3.5 billion USD). It has doubled in the last four years, and experts predict a further 100% increase to $7 billion in the next two to four years.

A major impediment to successful drug development is the complexity, cost, and scale of clinical trials. The detailed internal structure of clinical trial data can make conventional optimization difficult to achieve. Recent advances in machine learning, specifically graph-structured data analysis, have the potential to enable significant progress in improving the clinical trial design. TrialGraph seeks to apply these methodologies to produce a proof-of-concept framework for developing models which can aid drug development and benefit patients. In this work, we first introduce a curated clinical trial data set compiled from the CT.gov, AACT and TrialTrove databases (n=1191 trials; representing one million patients) and describe the conversion of this data to graph-structured formats. We then detail the mathematical basis and implementation of a selection of graph machine learning algorithms, which typically use standard machine classifiers on graph data embedded in a low-dimensional feature space. We trained these models to predict side effect information for a clinical trial given information on the disease, existing medical conditions, and treatment. The MetaPath2Vec algorithm performed exceptionally well, with standard Logistic Regression, Decision Tree, Random Forest, Support Vector, and Neural Network classifiers exhibiting typical ROC-AUC scores of 0.85, 0.68, 0.86, 0.80, and 0.77, respectively. Remarkably, the best performing classifiers could only produce typical ROC-AUC scores of 0.70 when trained on equivalent array-structured data. Our work demonstrates that graph modelling can significantly improve prediction accuracy on appropriate datasets. Successive versions of the project that refine modelling assumptions and incorporate more data types can produce excellent predictors with real-world applications in drug development.

Data science and artificial intelligence are two technologies that are transforming the world. While artificial intelligence powers data science operations, data science is not completely dependent on AI. Data Science is leading the fourth industrial revolution. This era saw massive amounts of data being generated by people on a daily basis and data science provides a way for businesses to capitalize on this available data. Our society now has become so data-driven that every major decision is a data-backed calculative move.

The NASSCOM Centre of Excellence for Data Science & Artificial Intelligence is an initiative of the Government of Karnataka, industry leaders & NASSCOM, established on a PPP (Public Private Partnership) model, to catalyze innovation & accelerate the AI ecosystem.

Podcast Episode 123 In this episode of the SuperDataScience Podcast, I chat with the unstoppable, Rico Meinl. You will learn the practical applications of AI and why is it important, know the different tactics on how to apply your passion while still learning, and listen to a discussion on how to setup an AI lab at an e-commerce ready company. If you enjoyed this episode, check out show notes, resources, and more at https://www.superdatascience.com/123