AI-Driven Biomolecular Modeling — Hands-on

Use state-of-the-art AI to predict, assess, and apply protein structures. Learn how to craft MSAs, select templates, interpret confidence metrics, run monomer/multimer jobs, and prepare AI models for downstream tasks like docking, design, and variant analysis—with reproducible, automated pipelines.

AI-Driven Biomolecular Modeling (AlphaFold, RoseTTAFold, ESMfold)

Help Desk · WhatsApp

Session Index

Session 1 — Foundations: MSAs, Templates & Metrics Session 2 — Running AlphaFold/RoseTTAFold Session 3 — ESMFold, Multimers & Refinement Session 4 — Mini Capstone: AI Model → Application

Session 1

Fee: Rs 6300

Foundations: MSAs, Templates & Metrics

MSA and template strategy for robust predictions

HHblits/JackHMMER MMseqs2 PDB templates

Confidence metrics and model selection

pLDDT PAE iptm/ptm (multimer)

Pitfalls: disorder, repeats, low-depth MSAs, domain splits

domain segmentation linkers

Session 2

Fee: Rs 8400

Running AlphaFold/RoseTTAFold

Local, server and cloud options; hardware & caching tips

GPU/CPU choices databases mgmt batch runs

Tuning inputs: MSAs, templates, recycles, ensembles

num_recycles num_ensembles template filters

Interpreting outputs, ranking, and preparing models for use

relaxation B-factor/pLDDT mapping

Session 3

Fee: Rs 11200

ESMFold, Multimers & Refinement

ESMFold for fast predictions; when MSA-free helps

LM-only inference throughput vs accuracy

Complexes & interfaces: AlphaFold-Multimer/RoseTTAFold2

chain pairing iptm scoring interface QC

Relaxation, cleanup & docking-readiness

amber_relax clash/geometry checks missing loops

Session 4

Fee: Rs 14000

Mini Capstone: AI Model → Application

Predict, assess and use a model for a downstream task (docking/design/variants)

Theory + Practical

Pipeline automation & provenance (batch runs, configs, logs)

Snakemake/Nextflow Conda/Mamba Quarto summary

Deliverables: figures, tables, README & manifest

PDF CSV/TSV checksums

PDF

AI-Driven Biomolecular Modeling Training | AlphaFold, RoseTTAFold, ESMFold, Confidence & Multimers

AI-Driven Biomolecular Modeling — Hands-on