CCGP 1 Nov. 2021

WGS UPDATE

Brad Shaffer, Erin Toffelmier, Courtney Miller

A collaborative effort to conserve California flora and fauna using conservation & landscape genomics of threatened, commercially exploited and ecologically important species.

For more info, go to:

ccgproject.org

California Conservation Genomics Project

WGS Numbers

​

• ~19,600 samples to be sequenced
• 75% of PIs responded to survey:
• 14,755 samples in hand (great)
• 5,255 have gDNA extracted (OK)
• 1,417 have library preps completed (not so great)

Numbers from the UCLA Mini-Core�(Dan Oliveira)

• Samples received (tissues & gDNA): 2232
• Samples extracted: 1462
• Samples completed & sequenced: 576

​

• Current capacity: ~200/week
• Anticipate ~ 8-10,000 total, or about 40 weeks total work.

​

GOAL: July 1, 2022

So, where are we at?

Our Goal:

​

• All reference genomes assembled and ready to go by ~1 May 2022

​

• All WGS completed by 1 July 2022

​

• That gives us our NCE year to complete variant calling and the first pass analyses

​

Today’s meeting:

​

• Erin updates on WGS options, opportunities

​

• Short funding update

​

• Breakout rooms: what is going well, what isn’t, and how can we help?

​

• Informatics team updates

​

• Reference genome team updates

CCGP Fall Webinar 2021

www.CCGProject.org

​

November 2, 2021

​

Reminders

1. Please fill out the Species Info Survey for NCBI BioProject Setup
• Two sentence species description, funding info, websites, etc.

https://airtable.com/shrgr2szvVrnrl0DW

​

• Information on all the recent updates can be found at: https://www.ccgproject.org/data

​

CCGP Fall Webinar 2021

Whole Genome Sequencing

​

​

​

​

Reminders and Updates

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Guidelines

• Number of samples per species (or genus): ~150, distributed across the range.

​

• Sequencing Depth: mean 10x raw depth across the genome
• Actual coverage, after removing organelles and duplicates, will be lower.
• Contact us with questions!

​

​

​

​

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Options

Library Prep Options

1. CCGP Mini-Core
2. DIY
3. UC Genomics Core

​

Sequencing Options:

• Must use a UC Facility
• QB3 Genomics at UC Berkeley will charge $7/Gb – for partial lanes or full lanes (Must mention CCGP in your submission)
• Contact QB3 Director, Christopher Hann-Soden (channsoden@berkeley.edu) for information

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Options

1. CCGP Mini-Core: www.ccgproject.org/wgs-option-3
• Tissue or gDNA
• SeqWell tagmentation based library preparation
• Sequencing at QB3 Genomics - $7/Gb, shared Novaseq lanes

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Options

2. DIY: Generate libraries yourself
• Must use a UC sequencing core
• QB3 Genomics at UC Berkeley:
• can pool your libraries those of other projects IF you use indexes they provide
• will charge $7/Gb – for partial lanes or full lanes (Must mention CCGP in your submission)
• Contact QB3 Director, Christopher Hann-Soden (channsoden@berkeley.edu) for information

​

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Options

3. UC Core generates libraries
• Most expensive option
• Unless you need a specific library preparation type, we recommend using the CCGP Mini-Core

www.ccgproject.org/resources-wgs

Whole Genome Sequencing - Timelines

An updated timeline:

Field work: Spring 2022

Benchwork: Sequencing completed by 1 July 2022

​

Special Note:

We can only guarantee the CCGP Mini-Core will be able to accept new submissions until ~ June 2022.

www.ccgproject.org/resources-wgs

Whole Genome Sequencing – Data Submission

Updates:

1. CCGP will submit short read data to NCBI on behalf of projects.
• Data will be embargoed automatically until June 30, 2023
• You may opt for earlier release, if needed

​

• We encourage submission of partial data sets

​

• Projects with sequence data submitted will receive reference genome priority status, if the genome is not already complete.

​

​

www.ccgproject.org/data-wgs-ingest

Whole Genome Sequencing – Data Submission

Two paths for sequence data submission:

1. Labwork and sequencing through the CCGP Mini-Core
• Relevant sample metadata is required in the submission
• We are provided with data download links from QB3
2. Submit sequence data directly to our Data Wrangler, Cade Mirchandani
• Fill out the metadata forms
• Email Cade with the forms, and download links from the sequencing core
• www.ccgproject.org/data-wgs-ingest

www.ccgproject.org/data-wgs-ingest

Questions?

Reminders

1. Please fill out the Species Info Survey for NCBI BioProject Setup
• Two sentence species description, funding info, websites, etc.

https://airtable.com/shrgr2szvVrnrl0DW

​

• Information on all the recent updates can be found at: https://www.ccgproject.org/data

​

CCGP Fall Webinar 2021

We will return at: 11:03

Reference Genome Related:

Courtney Miller: courtneymiller@g.ucla.edu

Merly Escalona: mmescalo@ucsc.edu

​

WGS Data Ingest and Bioinformatics:

Cade Mirchandani: cmirchan@ucsc.edu

Erik Enbody: erik.enbody@gmail.com

​

CCGP Mini-Core:

Dan Oliveira: danrodriguesolive@gmail.com

​

QB3 Genomics at UC Berekely:

Christopher Hann-Soden: channsoden@berkeley.edu

​

Everything else:

Brad Shaffer: brad.shaffer@ucla.edu

Erin Toffelmier: etoff@ucla.edu

Contacts

BREAKOUT ROOMS

ROOM 1 = Animals and invertebrates - Bench work discussion

ROOM 2 = Animals and invertebrates- Field work discussion

ROOM 3 = Plants - Bench work discussion

ROOM 4 = Plants - Field work discussion

So you’ve got a pile of reads,

now what?:

A big data saga

Erik Enbody, Cade Mirchandani, and Russ Corbett-Detig

UC Santa Cruz

Who we are

Russ Corbett-Detig

Project PI

UC Santa Cruz

Biomolecular Engineering

Erik Enbody

Postdoctoral Scholar

Evolutionary genomics and bioinformatics

Cade Mirchandani Assistant Specialist

Bioinformatics and snakemake expert

What we are doing

• Our role:
• Data archiving
• Data processing
• High-throughput variant calling pipeline
• Ensures reproducibility
• Uniform dataset for comparative analysis
• Reduces time and cost
• Delivers analysis-ready data to each PI

​

​

The CCGP is a

big data problem

• 225 species
• ~100 samples each
• = 19,600 genomes
• > ~40 trillion bases!
• Getting you your variant requires:
• Efficient code
• Lacking in most tools
• Big compute
• Most infrastructure is insufficient

Reads from your species

Your CCGP reference genome

Trim sequencing adaptors

fastp

Map reads to reference genome

BWA -mem

Remove PCR duplicates

sambamba

Intervals split by N’s

GATK4

​

Interval 1 variant calling

​

Interval 1 variant DB

​

Interval 1

joint genotype

​

Interval 1

Variant filters

​

Your favorite species’ genotypes (VCF)

About the genotype deliveries

• GATK “best practices” filters are annotated
• Data delivery
• No variants will be removed
• No samples will be removed
• QC is at your own discretion
• See QC plots supplied

Wuitchik, Shultz, Arnold, Mirchandani, Enbody, Corbett-Detig, Sackton: A reproducible workflow for variant calling & comparative population genomics in nonmodel organisms. In prep

​

QC Dashboard

CCGP Variant Calling Pipeline

Snakemake

Reads from your species

Your CCGP reference genome

Trim sequencing adaptors

fastp

Map reads to reference genome

BWA -mem

Remove PCR duplicates

sambamba

Intervals split by N’s

GATK4

​

Interval 1 variant calling

​

Interval 1 variant DB

​

Interval 1

joint genotype

​

Interval 1

Variant filters

​

Your favorite species’ genotypes (VCF)

Wuitchik, Shultz, Arnold, Mirchandani, Enbody, Corbett-Detig, Sackton: A reproducible workflow for variant calling & comparative population genomics in nonmodel organisms. In prep

​

QC Dashboard

Design constraints

Larger & more “complete” genomes take longer

• GATK is an “accepted method” but is optimized for humans
• GATK speed is limited by our ability to split up the genome

QC of read and pileups help detect issues with sequencing

Species 1

Species n

...

Variant QC dashboard example

https://erikenbody.github.io/qc_dashboard_interactive_EXAMPLE.html

CCGP Reference Genome Updates

November 1^st & 2^nd, 2021

Overview of submissions

• 140-150 species expected

​

• Lab component complete for majority (70%) of tissues received

​

• Data available for over half of the tissues processed

​

*35 have both

Reference Genome Notifications

• Phase 1: Species completes HMW DNA isolation step

​

• Phase 2: Draft assembly complete (contig-level)*

​

• Phase 3: Assembly completed scaffolding stage

​

• Phase 4: Genome complete and data uploaded to secure server

Updates

Reference genomes

Merly Escalona

mmescalo@ucsc.edu

@MerlyEscalona

Nov 1-2, 2021

General information

• Pipeline update
• HiFI + OmniC for contig assembly
• better phased blocks
• Modifications for faster turnaround
• identification of major misassemblies
• breaking of misjoins
• no rejoins or manual joins
• Mitochondrial genome: MitoHiFi (DToL)
• NCBI organization and metadata
• Species BioProjects
• Started with species with metadata available and assemblies ready to be submitted.
• General status of the assemblies
• And some details

Complete assemblies

Next in queue

• Arctostaphylos glauca
• Laterallus jamaicensis
• Arizona elegans
• Actinemys marmorata
• Rallus limicola
• Embiotoca jacksoni
• Aptostichus stephencolberti
• Juglans californica
• Haliotis cracherodii
• Semicossyphus pulcher
• Passerculus sandwichensis

• Final contamination screening and prep
1. Lynx rufus
2. Quercus engelmannii
3. Anthopleura sola
4. Thamnophis sirtalis
• Early stages (contigging/scaffolding)
• Artemisiospiza belli
• Melospiza melodia
• Mytilus californianus
• Nereocystis luetkeana

​

You’ll receive notification soon

with final metrics...

You’ll receive updates while genomes moves along the pipeline

CA Black walnut

Juglans californica

Black abalone

Haliotis cracherodii