<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/1b00be061f524e3ba446ba122ec24e99&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/1b00be061f524e3ba446ba122ec24e99-0acbc84a50b68156.gif</thumbnail_url><duration>284.544</duration><title>Running Boltz2 Protein Ligand Cofolding Pipeline</title><description>This Loom explains how to run Boltz code folding to co-fold a protein sequence with a ligand smile string for binding and activity prediction. It describes that the simulation places the ligand into the protein pocket tied to biological activity, using an example of an ATP-competitive kinase inhibitor binding at the hinge region. The results include confidence scores, affinity probability heuristics, and a predicted pIC50, with optional triage to MD using ProteinLoginMD and MMPBSA/MMGBSA for more robust scoring. The walkthrough then shows using RevBind and Boltz2 with a demo file, specifying inputs like AXLChainA and the smile string CSV or manual entry, creating and managing the pipeline from the central hub.</description></oembed>