2016 Workshops

Therefore, this BroadE provides an exciting opportunity for people to learn more about how and why these resources and datasets may be applicable to their own research. The session may also provide introductory information about how to access or get assistance using these resources. Attendees do not need computational experience. Additionally, subsequent BroadE workshops are being planned that will explore and demonstrate these individual tools and datasets in greater detail through hands-on training.

We encourage you to register to attend this new symposium sponsored and organized by BroadE and the Broad Cancer Program!

On FireCloud, users can upload their own analysis methods to workspaces or run the Broad Institute’s best practice tools and pipelines. FireCloud also includes tutorial workspaces as well as carefully curated open and controlled-access TCGA workspaces that users can clone.

The aim of this Workshop is to introduce analysts to the FireCloud environment via hands-on tutorials.

We will present this information in four parts:

• Background and Theory
• Visual Tools and Management
• Understanding “Value Added”
• The A3 Structured Problem Solving Approach

This workshop is designed to introduce the operational management methodology of dynamic work design, and to then provide several tools that can be applied across a diverse range of work environments for lab managers, project managers, and platform employees.

• Basics of image analysis.
• Obtaining measurements from objects.
• Basics of machine-learning for phenotype identification.

This hands-on introduction to CellProfiler will be followed by case-studies on HCS and cell-type classification. At the end of the workshop there will be a breakout session where attendees will receive guidance on analyzing their own image data. If you are curious about automating the analysis of your microscopy data or want to become familiar with "what's possible," come to the workshop and see what's new in CellProfiler for 2016.

Quantitative Proteomics in Biology, Chemistry and Medicine
The course will provide biologists, chemical biologists and clinicians with a working knowledge of the most relevant proteomic technologies and data analysis methods and will describe how these methods are being applied in a wide range of collaborative research at Broad.

The workshop consists of an all-day lecture series and two optional hands-on tutorials on the second day.

Day 1: November 7 — Lecture
This workshop will focus on the core steps involved in calling variants with Broad's Genome Analysis Toolkit (GATK), using best practices developed by the GATK team. The GATK development team and invited guests will give talks explaining the rationale, theory, and real-world applications of the GATK Best Practices. You will learn why each step is essential to the variant-calling process, what key operations are performed on the data at each step, and how to use the GATK tools to get the most accurate and reliable results out of your dataset.

Special lecture topics include somatic variant calling with MuTect2 and somatic copy number variant calling.

Day 2: November 8  — Hands-on Tutorials

One of two optional hands-on tutorials will be available to select participants. In each tutorial, the GATK team will guide participants in applying GATK tools on provided example data. The two sessions reflect different stages of research — data generation and evaluation of results.

The morning session is titled Joint Variant Discovery, and participants will come out knowing how to use the GVCF workflow and how to interpret results. 

The afternoon session is titled Variant Callset Filtering and Evaluation, and participants will come out knowing how to use tools to evaluate and filter a variant callset.

We will present a high-level introduction to computational meta’omics, highlighting the state-of-the-art in the field as well as outstanding challenges. This will include an introduction to the biological goals of typical meta’omic studies and the bioinformatic processes currently available to achieve them. 

Topics will include:

• Reference genome-based community composition and functional profiling
• Methods for constructing new genomic references by de novo assembly
• The challenges associated with precisely quantifying members of a microbial community
• Functional analysis of the gene families in a community
• The association of gene families with their source organisms
• The combination of gene families into pathways for metabolic profiling

We will conclude with an overview of the statistical challenges inherent to analyzing the compositional data arising in meta’omic studies.

Workshop attendees will gain hands-on experience with these analyses using bioBakery: a comprehensive platform for the analysis of shotgun meta’omic sequencing data. bioBakery includes tools for fast, accurate microbial taxonomic profiling (MetaPhlAn2), organism-specific functional profiling (HUMAnN2), identification/tracking of microbial strains (StrainPhlAn/PanPhlAn), and pattern discovery in microbial communities (MaAsLin, BAnOCC, and HAllA). Interspersed with lecture content, attendees will work through meta’omic analysis tutorials in Google Cloud instances of the bioBakery virtual machine.

This workshop will be run by members of the Huttenhower Lab of the Harvard T.H. Chan School of Public Health and Broad Institute

1. Introduction to CRISPR-Cas9 systems and basic genome editing methodologies, including guide design, in vitro and in vivo delivery, and assays to monitor editing
2. Guide to CRISPR-Cas9-based screening applications (both loss- and gain-of-function)
3. Overview of frontiers in genome editing, including therapeutic delivery modes, approaches to enhancing specificity, and novel RNA-guided RNA-targeting enzymes

Each module will be followed by ample time for questions to ensure that participants get detailed advice tailored to their specific needs.

• Basics of image analysis.
• Obtaining measurements from objects.
• Basics of machine-learning for phenotype identification.

This hands-on introduction to CellProfiler will be followed by case-studies on HCS and cell-type classification. At the end of the workshop there will be a breakout session where attendees will receive guidance on analyzing their own image data. If you are curious about automating the analysis of your microscopy data or want to become familiar with "what's possible," come to the workshop and see what's new in CellProfiler for 2016.

FireCloud is a cancer genome analysis platform built on a cloud computing environment (Google) with co-located TCGA data that was created and designed for the community of cancer researchers. FireCloud is modeled on Firehose, the cancer genome analysis platform built by the Getz lab at the Broad Institute, which supports both small groups and major projects (e.g., TCGA, GTEx). Firehose is used by both production managers for large-scale analysis and analysts for interactive analysis, curation and manual review of data for publication.

Like Firehose, FireCloud is built around "workspaces," which have robust security and access control and which hold data, tools, workflows, and results. FireCloud will hold best-practice tools and workflows that are currently in use at the Broad. In addition, FireCloud will host pre-loaded workspaces (e.g., TCGA-THCA workspace) holding data, pipelines, and results. These workspaces can be cloned and used to co-analyze user uploaded data. TCGA data — both protected and open access — will be co-located with FireCloud. Protected access data will require ERA Commons authorization and appropriate dbGaP permissions.

In this workshop, participants will learn how to:

• navigate the GenePattern environment
• analyze and visualize gene expression (including RNA-seq) and other genomic data
• create reproducible pipelines of their research
• create research narratives with the GenePattern Notebook environment

Along the way we will highlight best practices and share pointers to relevant resources, including web sites that describe laboratory methods, antibody validation resources, and public ChIP-seq data repositories. The intended audience includes everyone from curious people to post-docs who are interested in incorporating ChIP-seq methods in their research. 

FireCloud is a cancer genome analysis platform built on a cloud computing environment (Google) with co-located TCGA data that was created and designed for the community of cancer researchers. FireCloud is modeled on Firehose, the cancer genome analysis platform built by the Getz lab at the Broad Institute, which supports both small groups and major projects (e.g., TCGA, GTEx). Firehose is used by both production managers for large-scale analysis and analysts for interactive analysis, curation, and manual review of data for publication.

Like Firehose, FireCloud is built around "workspaces," which have robust security and access control, and which hold data, tools, workflows, and results. FireCloud will hold best-practice tools and workflows that are currently in use at the Broad. In addition, FireCloud will host pre-loaded workspaces (e.g., TCGA-THCA workspace) holding data, pipelines, and results. These workspaces can be cloned and used to co-analyze user uploaded data. TCGA data — both protected and open access — will be co-located with FireCloud. Protected access data will require ERA Commons authorization and appropriate dbGaP permissions.

Specific topics that will be covered are:

• Fine mapping to identify candidate causal variants in GWAS loci
• LD Score regression to distinguish confounding from polygenicity, and to estimate heritability and genetic correlation between traits
• Identity-by-descent mapping as an alternative to single-variant GWAS
• Pathway analysis to detect enrichment of association signals in biologically relevant gene sets

The workshop will combine lectures introducing each of these topics with demonstrations and hands-on practical — in particular, for performing LD Score and pathway analysis. Tutorial datasets will be provided for attendees to try out demonstrated tools. At the end of the workshop, participants should have an understanding of the different approaches that are available to interpret genetic association study results.

This seminar will highlight key concepts that are relevant for the implementation of Fragment Based Drug Discovery and Structure-Based Drug Design (SBD). Major topics include:

• Key factors that motivate FBDD and SBD
• Physicochemical properties and chemical space
• NMR methods for screening and structure-based design
• Other biophysical methods for screening and structure-based design
• Examples of success in the clinic 

Like Firehose, FireCloud is built around "workspaces," which have robust security and access control, and which hold data, tools, workflows, and results. FireCloud will hold best-practice tools and workflows that are currently in use at the Broad. In addition, FireCloud will host pre-loaded workspaces (e.g., TCGA-THCA workspace) holding data, pipelines, and results. These workspaces can be cloned and used to co-analyze user uploaded data. TCGA data — both protected and open access — will be co-located with FireCloud. Protected access data will require ERA Commons authorization and appropriate dbGaP permissions.