Gianluca Gazzola

Automated Discovery of Novel Drug Formulations Using Predictive Iterated High Throughput Experimentation

Background We consider the problem of optimizing a liposomal drug formulation: a complex chemical system with many components (e.g., elements of a lipid library) that interact nonlinearly and synergistically in ways that cannot be predicted from first principles. Methodology/Principal Findings The optimization criterion in our experiments was the percent encapsulation of a target drug, Amphotericin B, detected experimentally via spectrophotometric assay. Optimization of such a complex system requires strategies that efficiently discover solutions in extremely large volumes of potential experimental space. We have designed and implemented a new strategy of evolutionary design of experiments (Evo-DoE), that efficiently explores high-dimensional spaces by coupling the power of computer and statistical modeling with experimentally measured responses in an iterative loop. Conclusions We demonstrate how iterative looping of modeling and experimentation can quickly produce new discoveries with significantly better experimental response, and how such looping can discover the chemical landscape underlying complex chemical systems.

Coping with complexity: Machine learning optimization of cell‐free protein synthesis

Biological systems contain complex metabolic pathways with many nonlinearities and synergies that make them difficult to predict from first principles. Protein synthesis is a canonical example of such a pathway. Here we show how cell‐free protein synthesis may be improved through a series of iterated high‐throughput experiments guided by a machine‐learning algorithm implementing a form of evolutionary design of experiments (Evo‐DoE). The algorithm predicts fruitful experiments from statistical models of the previous experimental results, combined with stochastic exploration of the experimental space. The desired experimental response, or evolutionary fitness, was defined as the yield of the target product, and new experimental conditions were discovered to have ∼350% greater yield than the standard. An analysis of the best experimental conditions discovered indicates that there are two distinct classes of kinetics, thus showing how our evolutionary design of experiments is capable of significant innovation, as well as gradual improvement. Biotechnol. Bioeng. 2011;108:2218–2228.

Inter-cluster connectivity analysis for technology opportunity discovery

In today’s competitive business environment, the timely identification of potential technology opportunities is becoming increasingly important for the strategic management of technology and innovation. Existing studies in the field of technology opportunity discovery (TOD) focus exclusively on patent textual information. In this article, we introduce a new method that tackles TOD via technology convergence, using both patent textual data and patent citation networks. We identify technology groups with high convergence potential by measuring connectivity between clusters of patents. From such technology groups we select pairs of core patents based on their technological relatedness, on their past involvement in convergence, and on the impact of their new potential convergence. We finally carry out TOD by extracting representative keywords from the text of the selected patent pairs and organizing them into the basic description of a new invention, which the potential convergence of the patent pair might produce. We illustrate our proposed method using a data set of U.S. patents in the field of digital information and security.

Two-phase edge outlier detection method for technology opportunity discovery

This article introduces a method for identifying potential opportunities of innovation arising from the convergence of different technological areas, based on the presence of edge outliers in a patent citation network. Edge outliers are detected via the assessment of their centrality; pairs of patents connected by edge outliers are then analyzed for technological relatedness and past involvement in technological convergence. The pairs with the highest potential for future convergence are finally selected and their keywords combined to suggest new directions of innovation. We illustrate our method on a data set of US patents in the field of digital information and security.

An Introduction to the Analysis of Functional Magnetic Resonance Imaging Data

Functional magnetic resonance imaging (fMRI) is a brain imaging technology primarily used to investigate how cognitive processes affect neural activity. Due to its non-invasiveness and high spatial resolution, this technology has quickly become one of the most important research tools in cognitive neuroscience and has played a growing role in a number of clinical applications. The interpretation of the results of an fMRI experiment involves the analysis of massive amounts of noisy, complex, multivariate data, resolved both spatially and temporally. The extraction of information from this data is a difficult and articulated task, which relies on methodologies lying at the intersection between image processing, statistics, and machine learning. We here introduce the reader to the rich and diverse literature in the fascinating field of fMRI data analysis, providing an overview of its main challenges and of the most common approaches to overcome them.

Chapter 4 Evolutionary Design of a Model of Self-Assembling Chemical Structures

Abstract We introduce a new variant of dissipative particle dynamics (DPD) models that include the possibility of dynamically forming and breaking strong bonds. This model exhibits different forms of self-assembly processes; some like micelle formation involve only weak bonds, and others like the ligation of oligomers involve both weak and strong bonds. Complex self-assembly processes are notoriously difficult to design and program. We empirically demonstrate an evolutionary algorithm that optimizes self-assembly processes like micelle formation and template-directed ligation.