Distributed Visualization for Genomic Analysis
summary
Mango is a distributed genome browser that materializes requested regions from persistent storage (BAM/VCF or Apache Parquet via ADAM) into memory organized as an interval RDD backed by interval trees, enabling fast two‑dimensional range queries over ~2,500-sample variant data (terabytes). It uses data tiling and layered summaries to show raw reads up to sliding-window aggregates, runs on Spark for horizontal scaling, prefetches neighbor regions, and approaches interactive latency (~500 ms) once cluster cores exceed single-node limits; source on GitHub.
AI-written summary · about
chapters (10)
- 0:00Introduction to team
- 1:18What is genomic analysis
- 2:23From sequencing to variants
- 4:52Challenge of big genomic data
- 6:40Ideal genome browser goals
- 7:45Introducing Mango browser
- 11:17Optimizing for low latency
- 16:16Local performance comparison
- 17:18Scaling with distributed data
- 18:24Future improvements and Q&A
related talks
36mShow me your data!
46mExplorations in Interactive Visual Analytics
34mSearching for My Next Chart
16mVisualization and Analysis of Large Scale Datasets with Python
42mInteractive Data Analysis: Visualization and Beyond
01h 25mA practical guide to analysis and interactive visualization of massive datasets
03h 07mData of Unusual Size: Interactive Visualization
30mSeeing the needle AND the haystack: single-datapoint selection for billion-poin...