Introduction to Deep Learning with Caffe and Tensorflow

Deep Learning has emerged as an important new tool for a range of research applications. Emerging programs such as Caffe and Tensorflow are available at MSI and can be executed on Nvidia GPU resources. In this tutorial, MSI and Nvidia instructors will walk through an interactive exercise running some of these programs. Participants will be able to login to a node with a GPU to work through introductory exercises. We will also show how to run these programs in batch jobs on MSI systems.

Getting started with XSEDE resources

The NSF has a number of computational resources available to University staff and students as part of the XSEDE program. These computers have a lot in common with MSI systems, but there are a few important differences. This hands-on tutorial will introduce MSI researchers to multiple XSEDE resources. Participants will receive access to the Stampede 2 and Jetstream systems to run example jobs. This tutorial will cover the basics of policies, data movement, submitting jobs, and connecting to machines that are part of XSEDE.

Proteomics Mass Spectrometry Data Analysis

Introduction to common techniques for the analysis and quantitation of proteomics mass spectrometry ("mass spec") data.  Topics covered will include the concept of Peptide-Spectra Matching (PSM) versus de novo sequencing, as well as the application of commonly used algorithms such as Andromeda (Maxquant), Comet (Sequest), and Mascot.  Key parameters for different search engines will be discussed along with the statistical significance of their respective outputs.  Statistics for these data types - and how quantitation can be done for different experimental setups - will be discussed in the

Interacting with dbGaP data on Stratus

MSI has deployed a cloud service for research computing called Stratus. In its initial iteration, Stratus is designed expressly to satisfy the requirements set forth by the NIH Genomic Data Sharing (GDS) Policy for data from the Database of Genotypes and Phenotypes (i.e., dbGaP data). This tutorial introduces Stratus to users who wish to process dbGaP data at MSI, and gives them an interactive lesson on how to access the service, deploy their first virtual machines, and move data through multiple tiers of storage.

Quality Control of Illumina Data at the Command Line

Want to take your analysis to the next level? 

This hands-on tutorial will introduce new bioinformaticitions to the skills needed to start to build more complex analysis pipelines. This tutorial will take you though the steps needed to run standard quality control analysis on Illumina data using scripting to automate the analysis. 

In this tutorial you will be introduced to:

Basics of RNA-Seq Data Analysis - Lecture

This lecture will cover the basics of RNA-Seq experimental design and data quality assessment, followed by an overview of data analysis for the detection of differentally expressed genes.  Specific subtopics include:

Analysis of PacBio Sequencing Data Using SMRT Portal

This hands-on tutorial will cover installation and use of the SMRT portal at MSI to analyze PacBio sequencing data. The basics of full genome assembly and transcript assembly will be covered.  At the end of this tutorial, participants should be able to:

PacBio Sequencing - Lecture

This lecture will cover the special capabilities and use cases of PacBio sequencing as well as the basics of data analysis. Specific subtopics include:

  • Technology overview (physical basis of sequencing, pros and cons compared with other sequencing technologies)
  • De novo assembly applications (N50 and other assembly concepts, HGAP algorithm, diploid assembly)
  • IsoSeq transcriptome assessment (motivation, experimental procedure, biological applications, analysis approaches)
  • Visualization of PacBio data with new IGV features

Data Storage Systems and Data Analysis Workflows for Research

In this tutorial you will learn about the data storage systems available for academic research at the University of Minnesota. An overview of the kinds of storage systems that are available, policies for getting access to them, a comparison of their characteristics, and examples of how they can be accessed will be presented. You will also be given an overview of how the characteristics of UMN storage will impact the stability and throughput of various applications and workflows.


This one-day, hands-on workshop provides an introduction on how to write a parallel program using MPI and will help researchers write better and portable parallel codes for distributed-memory Linux clusters. The tutorial will focus on basic point-to-point communication and collective communications, which are the most commonly used MPI routines in high- performance scientific computation. In addition, the advantage of using MPI non-blocking communication will be introduced. Each session of the workshop will combine a lecture with hands-on practice.