Updated via the Open Science Framework

meeting-notes/2015-11-20_computational_pipeline.md

@@ -1 +1,67 @@
+# TERRAref Genomics Pipeline Meeting
 
+* Date: Friday, Nov 20, 2015
+* Time: 11:00-12:00 pm CST
+
+## Participants:
+
+* David Lebauer (TERRAref & TERRA cat 1, Illinois)
+* Vicor Jongeneel (HPCBio, Illinois)
+* Matt Hudson (Illinois)
+* Chris Fields (HPCBio, Illinois)
+* Noah Fahlgren (Danforth)
+* Rob Alba (Danforth)
+
+## Tenative:
+
+* Elodie Gazave (Cornell)
+* Pat Brown (TERRA cat 1, Illinois)
+
+### Regrets:
+
+* Mike Gore (Cornell)
+* Todd Mockler (Danforth)
+
+### Agenda
+
+* Review proposal terraref/reference-data#19
+* Determine roles / interests / contributions of each participating group
+
+
+## TERRA REF
+
+* 75 shared lines
+* 40 de novo sequences, 
+* 15 lines by PacBio. 
+* 400 resequences  (enough coverage with enough coverage to do assembly 
+
+
+Question: where to partition 15, 40, 400
+
+Coordinated germplasm for G x E analysis
+
+## TERRA MEPP
+
+Sequencing? Yet to be determined. Pending discussions like this.
+
+1. against existing reference genome
+2. re-analyze data with the PacBio lines (with a separate pipeline)
+
+
+## HPC Bio
+
+* ED: should spread 15 PacBiom through diversity of germplam + more coverage on more likely to be commercial lines
+* What can TERRA Ref do for the program? (and what are storage needs?)
+* What can HPCbio provide?
+
+
+
+  * TERRA MEPP
+  * HPCBio
+* Estimate data volumes (order of magnitude?) for each step
+* Define Use cases
+
+Need to figure out data flows for the entire project. 
+
+* **Victor**: makes sense to analyze data at same place it is generated. Then deposit both raw data and derived products in common repository.
+* **Chris Fields**: browsing and visualization