Main szg0314 channel https://video.igem.org/c/szg0314_channel/videos Tue, 20 Oct 2026 02:51:57 GMT https://validator.w3.org/feed/docs/rss2.html PeerTube - https://video.igem.org Main szg0314 channel https://video.igem.org/client/assets/images/icons/icon-1500x1500.png https://video.igem.org/c/szg0314_channel/videos All rights reserved, unless otherwise specified in the terms specified at https://video.igem.org/about and potential licenses granted by each content's rightholder. https://video.igem.org/feeds/podcast/videos.xml?videoChannelId=4045 <![CDATA[EdgeSci-THU-China: Lesson 6 (2026) [English]]]> https://video.igem.org/w/uMysupUQxpw6K3RCaXfo5V https://video.igem.org/w/uMysupUQxpw6K3RCaXfo5V Mon, 19 Oct 2026 14:08:59 GMT Lesson 6 moves from DDI prediction to investigation. It shows how multi-agent reasoning can organize evidence, handle uncertainty, build testable hypotheses, and connect computational predictions with wet-lab experiments. The lesson also introduces ECoPI as a practical evidence-to-experiment workflow.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 6 (2026) [English] Lesson 6 moves from DDI prediction to investigation. It shows how multi-agent reasoning can organize evidence, handle uncertainty, build testable hypotheses, and connect computational predictions with wet-lab experiments. The lesson also introduce... <![CDATA[EdgeSci-THU-China: Lesson 5 (2026) [English]]]> https://video.igem.org/w/kbpakCZfT4dPHfqNkSYyzb https://video.igem.org/w/kbpakCZfT4dPHfqNkSYyzb Mon, 19 Oct 2026 14:07:49 GMT Lesson 5 explores how biomedical knowledge can support drug interaction prediction when direct records are missing. Using EmerGNN, we follow how graphs connect drugs, select relevant paths, and combine evidence. We then examine prediction results, inspect candidate explanations, and distinguish useful model clues from biological mechanisms that need independent validation.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 5 (2026) [English] Lesson 5 explores how biomedical knowledge can support drug interaction prediction when direct records are missing. Using EmerGNN, we follow how graphs connect drugs, select relevant paths, and combine evidence. We then examine prediction results,... <![CDATA[EdgeSci-THU-China: Lesson 4 (2026) [English]]]> https://video.igem.org/w/96GMNrVZGhjdP5euJUgFNk https://video.igem.org/w/96GMNrVZGhjdP5euJUgFNk Mon, 19 Oct 2026 14:06:54 GMT Lesson 4 explores why DDI prediction becomes harder for emerging drugs. It introduces old–old, old–new, and new–new settings, explains why random splits can overestimate performance, and shows how distribution shift and DDI-Ben provide a more realistic way to evaluate new-drug generalization.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 4 (2026) [English] Lesson 4 explores why DDI prediction becomes harder for emerging drugs. It introduces old–old, old–new, and new–new settings, explains why random splits can overestimate performance, and shows how distribution shift and DDI-Ben provide a more real... <![CDATA[EdgeSci-THU-China: Lesson 3 (2026) [English]]]> https://video.igem.org/w/oiGxjdaC7T8VqG95WjVUCJ https://video.igem.org/w/oiGxjdaC7T8VqG95WjVUCJ Mon, 19 Oct 2026 13:48:22 GMT What’s inside a drug interaction prediction model? This lesson explores four approaches: rules, machine learning, graph models, and language models with agents. Through simple examples, we’ll discover what each method sees, what it assumes, and where it can go wrong.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 3 (2026) [English] What’s inside a drug interaction prediction model? This lesson explores four approaches: rules, machine learning, graph models, and language models with agents. Through simple examples, we’ll discover what each method sees, what it assumes, and wh... <![CDATA[EdgeSci-THU-China: Lesson 2 (2026) [English]]]> https://video.igem.org/w/eBiSqQBg4Abt7EoLN8EEGh https://video.igem.org/w/eBiSqQBg4Abt7EoLN8EEGh Mon, 19 Oct 2026 12:27:29 GMT You already know how drug–drug interactions work biologically. This lesson turns that knowledge into a prediction task a machine can learn: what counts as a sample, what to feed the model, what to ask it to predict, and how to judge the answer honestly.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 2 (2026) [English] You already know how drug–drug interactions work biologically. This lesson turns that knowledge into a prediction task a machine can learn: what counts as a sample, what to feed the model, what to ask it to predict, and how to judge the answer hon... <![CDATA[EdgeSci-THU-China: Lesson 1 (2026) [English]]]> https://video.igem.org/w/eCyoabaZFnUcFShN54xzF2 https://video.igem.org/w/eCyoabaZFnUcFShN54xzF2 Mon, 19 Oct 2026 12:17:41 GMT Welcome to Lesson 1! Let's explore how drug-drug interaction (DDI) occur within the body and why it endanger patient safety. We’ll examine the key biological interaction sites and analyze PK and PD mechanisms accompanied by real-world cases to build a foundation for AI-driven risk prediction.

https://2026.igem.wiki/edgesci-thu-china

]]> Main szg0314 channel Education nonadult EdgeSci-THU-China: Lesson 1 (2026) [English] Welcome to Lesson 1! Let's explore how drug-drug interaction (DDI) occur within the body and why it endanger patient safety. We’ll examine the key biological interaction sites and analyze PK and PD mechanisms accompanied by real-world cases to bui...