Learning with Kernels on Graphs: DAG-based kernels, data streams and RNA function prediction.

In many application domains data can be naturally represented as graphs. When the application of analytical solutions for a given problem is unfeasible, machine learning techniques could be a viable way to solve the problem. Classical machine learning techniques are defined for data represented in a vectorial form. Recently some of them have been extended to deal directly with structured data. Among those techniques, kernel methods have shown promising results both from the computational complexity and the predictive performance point of view. Kernel methods allow to avoid an explicit mapping in a vectorial form relying on kernel functions, which informally are functions calculating a similarity measure between two entities. However, the definition of good kernels for graphs is a challenging problem because of the difficulty to find a good tradeoff between computational complexity and expressiveness. Another problem we face is learning on data streams, where a potentially unbounded sequence of data is generated by some sources. There are three main contributions in this thesis. The first contribution is the definition of a new family of kernels for graphs based on Directed Acyclic Graphs (DAGs). We analyzed two kernels from this family, achieving state-of-the-art results from both the computational and the classification point of view on real-world datasets. The second contribution consists in making the application of learning algorithms for streams of graphs feasible. Moreover,we defined a principled way for the memory management. The third contribution is the application of machine learning techniques for structured data to non-coding RNA function prediction. In this setting, the secondary structure is thought to carry relevant information. However, existing methods considering the secondary structure have prohibitively high computational complexity. We propose to apply kernel methods on this domain, obtaining state-of-the-art results.

Purple bacteria for the biotechnological valorisation of agro-industrial streams

Waste management worldwide has received increasing attention from global policies in recent years. In particular, agro-industrial streams represent a global concern due to the huge volumes generated and a high number of residues, which produce an environmental and economic impact on the ecosystem. The use of biotechnological approaches to treat these streams could allow the production of desirable by-products to be reinjected into the production cycle through sustainable processes. Purple phototrophic bacteria (PPB) are targeted as microorganisms capable to reduce the pressure of agro-industrial streams on environmental issues, due to their metabolic versatility (autotrophic and/or heterotrophic growth under different conditions). This Ph.D. research project aims to assess the effectiveness of PPB cultivation for industrial streams valorisation in the applications of biogas desulfurization and microbial protein production. For these purposes, the first part of the present work is dedicated to the cultivation of purple sulfur bacteria (PSB) for biogas streams upgrading, cleaning biogas from sulfur compounds (H2S), and producing elemental sulfur (S0), potentially suitable as a slow-release fertilizer. The second part of the thesis, instead, sees the application of purple non-sulfur bacteria (PNSB) on streams rich in organics, such as molasses, generating biomass with high content of proteins and pigments, useful as supplements in animal feed. The assessment of the main metabolic mechanisms involved in the two processes is evaluated at a laboratory scale using flasks and a photobioreactor, to define the consumption of substrates and the accumulation of products both in the autotrophic (on biogas) and in heterotrophic grow (on molasses). In conclusion, the effectiveness of processes employing PPB for a sustainable valorisation of several agro-industrial streams has been proved promising, using actual residues, and coupling their treatments with the production of added-value by-products.