July 2022 Summaries

In the blog post by Jure Bajic, the author details the process of adapting Memgraph's software to support multi-architecture Docker images, particularly for AMD64 and ARM64 architectures, in response to the prevalence of Apple's M1 chip. The explanation includes step-by-step instructions on setting up the Docker CLI extension buildx on Ubuntu 20.04, which enables the building of images for multiple architectures, as well as the installation of a builder instance from the tonistiigi/binfmt repository to support additional platforms. The article also discusses the creation and testing of a Dockerfile tailored to different architectures using the buildx tool, highlighting the importance of automation via GitHubActions for efficiency in building and pushing Docker images to Dockerhub. Additionally, Bajic addresses the complexities of architecture naming conventions, noting the distinctions between Debian and RPM packages and their corresponding suffixes, ultimately aligning Docker image conventions with those used for Debian packages. The author concludes by emphasizing the benefits of using CI solutions like GitHubActions for releasing multi-architecture products, while also recognizing the value of manual configuration for deeper technological understanding.

The 2022 summer internship program at Memgraph has commenced, welcoming six students from the Faculty of Science and the Faculty of Electrical Engineering and Computing to gain hands-on experience with Memgraph products. This year's program features two main projects: MAGE, an open-source library for graph analytics where interns will focus on research and algorithm development, and MagicGraph, a project aimed at developing web applications using Memgraph to solve real-world problems. To enhance the remote experience, a casual meet-up was held to introduce students to their mentors and the Memgraph team, with plans for a concluding event to showcase their work. Interns expressed enthusiasm about the opportunity to work on cutting-edge topics like graph neural networks, enterprise-grade coding, and web development, while learning about the power and construction of graph databases. Although this year's application period is over, students are encouraged to stay tuned for future opportunities with Memgraph.

The Graph Data Zagreb meetup season concluded on July 7, 2022, with an engaging finale featuring Michaël Ughetto, a graph data scientist from AstraZeneca, who discussed the use of a Biomedical Knowledge Graph (BIKG) in accelerating drug discovery. The season comprised four meetups that attracted over 240 graph enthusiasts, covering topics such as Graph Tech Landscape and GitHub Repository Vulnerabilities Detection, and included a presentation on graph analytics using a Game of Thrones dataset. The BIKG, an internal tool at AstraZeneca, integrates data from various sources to aid drug development through machine learning and analytics, offering insights into new drug targets and existing drug repurposing. The organizers announced plans for a second season starting in September, inviting feedback and engaging with their community on platforms like Discord and Facebook.

The blog post explores the integration of ChemicalX, a graph machine learning library, with Memgraph to predict drug interactions in the pharmaceutical industry, focusing on drug synergy, polypharmacy side effects, and drug-drug interactions. It highlights how graph technology is used in drug discovery to model interactions between different drugs and entities like proteins and genes, providing a foundation for data-driven predictions of drug effects. The article details the process of building a synergy prediction module using Memgraph's ecosystem and ChemicalX's neural network architectures, illustrating how the integration can enhance the prediction of drug interactions and ultimately improve the efficiency of drug development. It emphasizes the potential of graph databases to handle complex datasets in drug research and invites readers to engage with the Memgraph community for further exploration and development.