SMRT-Portal Exercises

SMRT-Portal ExercisesJ Fass

UCD Genome Center Bioinformatics CoreThursday April 16, 2015

Running SMRT-Portal in AWS

see PacBio documentation

We’ll be running a virtual machine (VM) in the Amazon Web services “Cloud” (a server farm somewhere in the region you’ve selected). On this VM is a web server, serving you pages created by the SMRT-Portal application.

UC Davis Genome Center | Bioinformatics Core | J Fass SMRT-Portal Exercises 2015-04-16

https://github.com/PacificBiosciences/Bioinformatics-Training/wiki/%22Installing%22-SMRT-Portal-the-easy-way---Launching-A-SMRT-Portal-AMI


Launch an m3.2xlarge instance using ami-953fddd1.

Generate or re-use a key pair - you will need it!

Once running, find the public IP address (#.#.#.#), and open a browser tab with the URL:#.#.#.#:8080/smrtanalysis … or … #.#.#.#:8080/smrtportal



On a “vanilla” PacBio SMRT-Portal instance (U.S. East / N. Virginia), you would need to create one administrator account. This AMI already has one, but feel free to change the password, add non-admin accounts, etc.

user: administratorpwd: 5MRT-P0rtal

Note: pwd = >0 symbols, >0 numbers, >8 characters



Log in as administrator (special user), then create separate accounts if desired.


How I imported 8 SMRT Cells (E coli)


SSH to AWS instance

ssh -i ~/.ssh/yourKey.pem [email protected]

ssh commandoption block (supplies private key in this case)destination (username@computername)

… (or use PuTTY) …


PacBio Public Datasets

https://github.com/PacificBiosciences/DevNet/wiki/Datasets

look for “Data supporting publications” …






look for “Data supporting publications” … look for the first MG1655 xml & bas.h5 files …




Enter “dropbox” directory

cd /opt/smrtanalysis/userdata/inputs_dropbox

cd commanddestination directory


Pull in data

mkdir MG1655

cd MG1655

wget [xml file link]

mkdir Analysis_Results

cd Analysis_Results

wget [bas.h5 file link, + bax.h5’s if present]

commanddirectory / destination / source


Import SMRT Cell data

Back in SMRT Portal, click through “Home” (upper left), then “Import and Manage” (third image), then “Input SMRT Cells.”


Import another SMRT Cell (exercise)


SSH to AWS instance

ssh -i ~/.ssh/yourKey.pem [email protected]

ssh commandoption block (supplies private key in this case)destination (username@computername)

… (or use PuTTY) …




look for “E. coli size selected 20kb library” …





Find the SMRT Cell data files “tarball,” and copy the link (don’t download; you’ll break our wireless!).


Feeding Data to the SMRT-Portal

Back in a shell (terminal) on your instance, navigate to SMRT-Portal’s input dropbox.

cd /opt/smrtanalysis/userdata/inputs_dropboxwget [link]mkdir Ecoli20kbcd Ecoli20kbmkdir Analysis_Resultstar -xzvf ecoliK12.tar.gz




Back in SMRT Portal, click through “Home” (upper left), then “Import and Manage” (third image), then “Input SMRT Cells.”



Via Home, Import and Manage, and [Import] SMRT cells, get to import page. Select directory, and Scan.


HGAP Assembly

Running HGAP


Click Design Job, then Create New, (deal with the design wizard - I usually select “display all protocols”). You should see 9 SMRT Cells available (we just imported the 9th).

Running HGAP


Select the “RS_HGAP_Assembly.3” Protocol from the drop-down menu, enter name and (if desired) comments, select 20kb cell and click right arrowhead to add cell to the job you’re designing, then Save and Start!

Running HGAP


early results just assess reads, subreads ...

Running HGAP


Final results include pre-assembly, realigned reads, etc.

HGAP output


Find the Polished Assembly Fasta link, right-click and Save link as … (to avoid troublesome name).

HGAP output


Notice the BAM and BAI links; these allow you to view the original reads aligned back to the assembly (e.g. in IGV).

Check assembly via homology


Using Mauve, we’ll align our assembled genome to the trusted E. coli K-12 MG1655 reference assembly, from GenBank (link).

http://www.ncbi.nlm.nih.gov/nuccore/556503834



Launch Mauve, then select File → Align with progressiveMauve. Then Add Sequence (click to add GenBank reference, then our assembly), click Align (and add a place to save output).



(see Mauve site for details on viewer, etc. … we’ll explore during Workshop)

http://darlinglab.org/mauve/mauve.html

Check for circularity (if appropriate)


Launch Gepard, Select file specifying the polished genome assembly twice (once for horizontal, once for vertical), then create dotplot.



Looks fine, right? But the overlaps will be on the size scale of the reads … not visible at this scale.



Use the Advanced mode, Plot tab, to specify the first ~20kb on the horizontal, and the last ~20kb on the vertical. Then Update dotplot.


Alignment / Resequencing Protocols

Align to your own reference

In SMRT-Portal, go Home, then Import and Manage, then reference sequences. Select New to upload our down loaded reference (note there’s also a Scan option - upload first to /opt/smrtanalysis/userdata/references_dropbox/).


Align to your own reference

Design a job using the same reads, and the RS_Resequencing.1 protocol. Specify your uploaded reference sequence, save, and start the job. (I’m using E albertii in this case, RefSeq id NZ_CP007025.1)



Viewing Read Alignments with IGV

SMRT-Portal Exercises

Documents