Bioinformatics Newsletter 1: 22 Jan 2016

Announcements:

East Coasters – Stay Warm! The blizzard LOOMS! Find some of your favorite scientist’s publications and cozy up with a hot beverage, I know I will!

Latest and greatest in the Lit

Escobar-Zepeda et al., 2015. The road to metagenomics: from microbiology to DNA sequencing technologies and bioinformatics. Front in Genet http://journal.frontiersin.org/article/10.3389/fgene.2015.00348/full

Brierley et al., 2016. Quantifying global drivers of zoonotic bat viruses: a process-based perspective. Amer Naturalist. http://www.journals.uchicago.edu/doi/abs/10.1086/684391?journalCode=an

Borucki et al., 2016. Middle east respiratory syndrome coronavirus intra-host populations are characterized by numerous high frequency variants. PLOS ONE. http://dx.doi.org/10.1371/journal.pone.0146251

Didelot et al., 2016. Within-host evolution of bacterial pathogens. Nature Reviews http://www.nature.com/nrmicro/journal/vaop/ncurrent/abs/nrmicro.2015.13.html

Metcalf et al., 2016. Microbial community assembly and metabolic function during mammalian corpse decomposition. Science. http://science.sciencemag.org/content/351/6269/158.full

Gill et al., 2016. Understanding past population dynamics: Bayesian coalescent-based modeling with covariates. arXiv http://arxiv.org/abs/1601.05078

Ziegenhain et al., 2016. Comparative analysis of single-cell RNA-sequencing methods. bioRxiv http://biorxiv.org/content/early/2016/01/05/035758

Parekh et al., 2016. The impact of amplification on differential expression analyses by RNA-seq. bioRxiv http://biorxiv.org/content/early/2016/01/15/035493

Kim et al., 2016. Host specific and segment-specific evolutionary dynamics of avian and human influenza A viruses: A systematic review. PLOS ONE. http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0147021

Scientific Internet Chatter

“CDC Zika press conference: Audio and Transcript” (Avian Flu Diary) http://afludiary.blogspot.com/2016/01/cdc-zika-press-conference-audio.html

“Will Zika become the 2016 NTD of the Year?” (PLOS Blogs) http://blogs.plos.org/blog/2016/01/07/will-zika-become-the-2016-ntd-of-the-year/

“Almost everything you wanted to know about Illumina HiSeq 4000” (Core Genomics) http://core-genomics.blogspot.com/2016/01/almost-everything-you-wanted-to-know.html

“Databases in Bioinformatics” (YouTube) Hours 1 and 2: https://www.youtube.com/watch?v=CHwTpBQBwbM and https://www.youtube.com/watch?v=vOcLJkfbX60

“BIOM23: a 16S Practical” (Loman Lab) http://lab.loman.net/2016/01/22/biom25-16s-practical/

“Paper summary: fast and accurate single-cell RNA-seq analysis using equivalence class counts” (NextGenSeek) http://nextgenseek.com/2016/01/paper-summary-fast-and-accurate-single-cell-rna-seq-analysis-using-equivalence-class-counts/

Software Notes:

Blischak, Daveport and Wilson. 2016. A quick introduction to version control with Git and Github. PLOS Comp Biol http://dx.doi.org/10.1371/journal.pcbi.1004668

“RNA-seq Analysis Pipeline – TUXEDO protocol” (YouTube) http://www.rna-seqblog.com/rna-seq-analysis-pipeline-tuxedo-protocol/

Bens et al., 2016. FRAMA: from RNA-seq data to annotated mRNA assemblies. BMC Genomics.  http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4712544/

“BroadE- GATK/Mapping and processing RNAseq” (RNAseq Blog) http://www.rna-seqblog.com/broade-gatkmapping-and-processing-rnaseq/

Medina et al., 2016. HPGAlign: Highly sensitive and ultrafast read mapping for RNA-seq analysis. DNA Res http://dnaresearch.oxfordjournals.org/content/early/2016/01/05/dnares.dsv039.long

Korthauer et al., 2016. scDD: A statistical approach for identifying differential distributions in single-cell RNA-seq experiments. bioRxiv http://biorxiv.org/content/early/2015/12/29/035501

Gu et al., 2015. Depletion of abundant sequences by hybridization (DASH): Using Cas9 to remove unwanted high-abundance species in sequencing libraries and molecular counting applications. bioRxiv. http://dx.doi.org/10.1101/031708

101 Fun Bioinformatics Facts!

#14 Velvet is so named because Daniel Zerbino wore velvet gloves when coding it (via Nick Loman)

Dusting off the blog site...Newsletter time

Greetings and Salutations oh Minions of Bioinformatics...

So in keeping with the premise of this blog as you'll notice - I haven't exactly gone anywhere or done anything meriting a hash on this site. Well that's not totally true - I did once again give my Phylogenetics/Sequence Analysis workshop in January 2015 - this time in Peru and I am shamefully delinquent on posting the materials for that but it's on the list.

My excuse? Well I had just found out I was pregnant. That being said I spent the following 9 mos alternately excited, busy attempting to thwart morning sickness and keep up with work and fretting about 10 pound baby head nightmares. Then I was on maternitiy leave. I just recently returned from leave and have a healthy gorgeous baby girl who lights me up no matter how freaking tired I am...and I am tired.

Thus! I have attended no conferences although my work has appeared at conferences with others presenting. Nor have I attended any workshops. I did have two papers accepted (Huzzah!) and 2 more that'll go into submission before the end of the year - so it hasn't been a total intellectual loss.

As part of my return I emailed my Bioinformatics Enthusiasts listserve with the monthly offering in the form of a Bioinformatics Newsletter that I've been putting together and sharing for the past 3 years. Why haven't I posted it here? Because for the past 3 years the institution I work for has blocked this site as a 'blog' site. I don't know why they suddenly unblocked it but I'll take it! And so here I will post my monthly one-hit-wonder for those that are interested in what I find interesting in the world of virolgy and bioinformatics.

Many thanks to my utterly fabulous colleague Nick Loman for encouraging me to start posting these online - here's the shameless plug to his site: Go Here I am Nick Loman's Awesome Lab Site. If I were on twitter right now I'd be hash-tagging that sentence with #scigroupie

Many thanks to my tens of followers of this blog, I hope you find it's content useful and now in the stead of all the conferences and workshops I am not currently attending - 

I shall post my newsletter instead.

 Cheers!

Announcements:

Happy Holidays!

Online Short Course:

Pathogen evolution, selection and immunity http://bedford.io/projects/sismid/

Latest and greatest in the Lit

Darriba, Flouri and Stamatakis. 2015. The state of software in evolutionary biology. bioRxiv. http://dx.doi.org/10.1101/031930     

Khang and Lau. 2015. Getting the most out of RNA-seq data analysis. PeerJ. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4631466/  

Wang et al., 2015. Reemergence of autochthonous transmission of dengue virus, Eastern China, 2014. EID http://wwwnc.cdc.gov/eid/article/21/9/15-0622_article

Duong et al., 2015. Asymptomatic humans transmit dengue virus to mosquiotes. PNAS. http://www.pnas.org/content/112/47/14688.abstract

Jennifer Doody. 2015. Ebola outbreak: a system that failed. Harvard Gazette. http://news.harvard.edu/gazette/story/2015/11/ebola-outbreak-a-system-that-failed/

Lau et al., 2015. A systematic Bayesian integration of epidemiological and genetic data. PLoS Comp Biol http://dx.doi.org/10.1371/journal.pcbi.1004633

Bowers et al., 2015. Impact of library preparation protocols on the metagenomic reconstruction of a mock microbial community. BMC Genomics. http://www.biomedcentral.com/1471-2164/16/856

Worby, Lipsitch and Hanage. 2015. Shared genomic variants: identification of transmission routes using pathogen deep sequencing data. bioRxiv http://dx.doi.org/10.1101/032458

Bedford et al., 2015. Global circulation patterns of seasonal influenza viruses vary with antigenic drift. Nature. http://www.nature.com/nature/journal/v523/n7559/full/nature14460.html

Scientific Internet Chatter

Tutorial: RNA-seq differential expression & pathway analysis with Sailfish, DESeq2, GAGE and Pathview. (Getting Genetics Done) http://www.gettinggeneticsdone.com/2015/12/tutorial-rna-seq-differential.html

A computational pipeline for cross-species analysis of RNA-seq data using R and Bioconductor. (RNAseq Blog) http://www.rna-seqblog.com/a-computational-pipeline-for-cross-species-analysis-of-rna-seq-data-using-r-and-bioconductor/

How to extract FASTQ from the new MinION FAST5 format using poRe. (Opiniomics) https://biomickwatson.wordpress.com/2015/11/23/how-to-extract-fastq-from-the-new-minion-fast5-format-using-pore/

“Do demons dreams of phylogeny packages?” (Omics Omics) http://omicsomics.blogspot.com/2015/11/do-demons-dream-of-phylogeny-packages.html

“The five habits of bad bioinformaticians” (Opiniomics) https://biomickwatson.wordpress.com/2015/11/16/the-five-habits-of-bad-bioinformaticians/

“DENGVAXIA®, worlds first dengue vaccine, approved in Mexico” (Sanofi Pasteur) http://sanofipasteur.com/en/articles/dengvaxia-world-s-first-dengue-vaccine-approved-in-mexico.aspx

“MinION and time to result” (Omics Omics) http://omicsomics.blogspot.com/2015/12/minion-and-time-to-result.html

“Oxford Nanopore’s Software Side.” (BioIT World) http://www.bio-itworld.com/2015/11/19/oxford-nanopores-software-side.html

Software Notes:

Lin and Liao. 2015. Evaluation and validation of assembling corrected PacBio long reads for microbial genome completion via hybrid approaches. PLoS ONE. http://dx.doi.org/10.1371/journal.pone.0144305  

Peabody et al., 2015. Evaluation of shotgun metagenomics sequence classification methods using in silico and in vitro simulated communities. BMC Bioinformatics. http://www.biomedcentral.com/1471-2105/16/363

Borght et al., 2015. QQ-SNV: single nucleotide variant detection at low frequency by comparing the quality quantiles. BMC Bioinformatics. http://www.biomedcentral.com/1471-2105/16/379

101 Fun Bioinformatics Facts (Link)!

#10 – The HGAP assembler is actually an elaborate front-end hiding 

three thousand slave labourers all running GAP4!

 

You can also find a totally exhaustive listing of my internet bioinformatic trolling spoils at http://www.scoop.it/u/mel-melendrez-vallard where I link to research in general virology, dengue, influenza, bioinformatic software that's published or coming out as well as post job openings and trainings I come across so feel free to hit that up if you desire more bioinformatic madness.

Want more Training in Bioinformatics?

More Training Opportunities, courtesy of Dr. Stephen Turner...

Dr. Stephen Turner runs the Getting Genetics Done Blog which has a lot of great posts regarding bioinformatics; general news, software developments and training. It's linked also over to the left as one of the blogs I follow.

He has compiled an extensive listing of Bioinformatics Workshops and Trainings which includes training programs, MOOCs (online courses/modules), workshops, short courses, literature, recommended reading and useful blogs.

It's worth a look.

Cheers,
Mel

---

Melanie Melendrez, Ph.D.

Contractor--CNTS

Chief, Bioinformatics, Lead Scientist

Walter Reed Army Institute of Research

Viral Diseases Branch

Silver Spring, MD 20910

melanie.c.melendrez.ctr@us.army.mil

Evomics 2014 Workshop!

Looking for the blog on the Workshop on Genomics for 2014?

Seems the workshops organizers enjoyed my blogging shenanigans so much last year they've asked me to blog from the Evomics website this year.

First blog post is up...

"Evomics 2014...ready, set go!"

If you'd like to relive the guts and glory from last year, the posts that received the highest traffic from the workshop last year are linked below!

• Comparison charts for NGS platforms 2013: a tangent I included during the workshop last year that got a surprisingly high number of pageviews.
• The wrap up from last year's workshop with highlights
• The Modern Genome Sequencing preparation blog
• Konrad's talk on Short Read alignment

and

• Konrad again..."So you want to be an NGS sequencer eh?" from day 2 where he discusses the latest and greatest in the technology.

Feel free to peruse others tagged as workshop on genomics or WoG and keep your eye on the Evomics blog for this years workshop rundown and roundup!

Ready, set...

GATK Best Practices Workshop: Variant Calling

GATK Best Practices Workshop
Variant Calling

"Examining the evidence for variation from reference via Bayesian inference"

There are two essential approaches to finding genetic variation that's 'real'.

1. Initial approach which is very fast and uses and independent base assumption
2. Evolved approach which is more computationally intensive and involves local de novo assembly of the variable region.

There were two variant callers discussed: The Unified Genotypes and the Haplotype Caller

Unified Genotyper (UG)

• Calls SNPs and indels separately by considering each variant locus independently
• Determine possible SNP and indel alleles
• Compute likelihoods of data given genotypes
• Compute allele frequency distribution to determine most likely allele count, omit a variant call if it's determined that it should be omitted.
• Assign genotypes to samples
• Accepts any 'ploidy'
• Can do pooled calling
• You need high sample numbers
• Remember you have to run indel realignment per the previous blog, it's required by UG. 

Bayesian modeling is used for SNP and indel calling, you can see all the modeling action here (this is an older presentation but you can download the new one per the forum link).

• Inference: What is genotype G given sample read data D
• Calculate (Bayes' rule) the probability of each possible genotype (G)
• Assumes reads are independent
• Relies on the likelihood function to estimate probability of sample data given a 'proposed' haplotype
• Considers the 'pileup' of bases and their associated quality scores
• Only considers "Good Bases"
• Good bases satisfy the minimum requirement for base quality, mapping read quality and pair mapping quality
• The prior on Bayesian inference is or was tuned using human data...if you are using a different data set you will want to tune your prior differently.
• Indels are more involved because the number of possibilities increases dramatically.
• In the end...you get simultaneous estimation of allele frequency, the probability that a variant exists and the assignment of genotypes to each sample

credit: http://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-0 (may not be most recent, search forum for most recent slides from Oct 2013 workshop)

Now that we've gone over Unified Genotyper...let me tell you about the program that will eventually replace it...doh!

Haplotype Caller

• Calls variants by local de novo assembly
• More accurate especially for indels

1. Propose haplotypes with loci de novo assembly using de Bruijn graphs
2. Evaluate haplotypes with Pair HMM

credit: http://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-0 (may not be most recent, search forum for most recent slides from Oct 2013 workshop)

As a note: If you invoke minPruning this will improve performance as it will 'toss' low confidence haplotypes.

These variant callers give you 'raw' calls...so more work is needed to assess the quality of the variant calls...

Variant Quality Score Recalibration (VQSR)

• Mutation callers typically 'cast a wide net' VQSR helps us narrow that 'net' to what is most likely 'real'.
• Requires A LOT of data, if you have a small number of variants or samples VQSR will not help your analysis.
• Allows analyst to trade off sensitivity and specificity depending on their project goals. (you can control this).
• Builds a model of what true genetic variation looks like and allows for rank-ordering of variants based on their likelihood of being real.
• Some assumptions and notes:
• Each variant has a diverse set of statistics associated with it (variant annotations)
• Real variants tend to cluster together
• Clusters tend to be gaussianly distributed (the GATK team noted this over many runs on many datasets)
• Uses a gaussian mixture model to fit the data
• Tries to find the fewest (smallest number) of clusters that explain the data.

credit: http://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-0 (may not be most recent, search forum for most recent slides from Oct 2013 workshop)

• First you will build your model (VariantRecalibrator) then apply filters and write the new annotated VCF (ApplyRecalibration) 
• Recalibrate your SNP calls and indels separately...how?
• In your first analysis do the SNPs
• In the second analysis use the recalibrated VCF and do indels (indel mode).

credit: http://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-0 (may not be most recent, search forum for most recent slides from Oct 2013 workshop)

Note: Plots require the R statistical package be installed...

credit: http://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-0 (may not be most recent, search forum for most recent slides from Oct 2013 workshop)

GATK definitely provides a step by step framework that should get you from raw data to variant calls pretty seamlessly. Remember though that many of their parameters and assumptions are based on their extensive work on human genome projects and may not be necessarily applicable to your bacterial or viral genome project.

That being said they are a receptive group and questions on the forum are welcome.

Additionally, there are variety of variant callers out there...for instance Mike Zody (also from the Broad Institute) has been involved with work on V-Phaser, V-Phaser 2 and V-Profiler; which are specialized callers for viral population data:

• Henn MR, Boutwell CL, Charlebois P, Lennon NJ, Power KA, Macalalad AR, Berlin AM, Malboeuf CM, Ryan EM, Gnerre S, Zody MC, Erlich RL, Green LM, Berical A, Wang Y, Casali M, Steeck H, Bloom AK, Dudek T, Tully D, Newman R, Axten KL, Gladden AD, Battis L, Kemper M, Zeng Q, Shea TP, Gujja S, Zedlack C, Gasser O, Brander C, Hess C, Gunthard HF, Brumme ZL, Brumme CJ, Bazner S, Rychert J, Tinsley JP, Mayer KH, Rosenberg E, Pereya F, Levin JZ, Young SK, Jessen H, Altfeld M, Birren BW, Walker BD, Allen TM(2012) Whole Genome Deep Sequencing of HIV-1 Reveals the Impact of Early Minor Variants Upon Immune Recognition During Acute Infection. PLoS Pathogens 8(3): e1002529
• Macalalad AR, Zody MC, Charlebois P, Lennon NJ, Newman RM, Malboeuf CM, Ryan EM, Boutwell CL, Power KA, Brackney DE, Pesko KN, Levin JZ, Ebel GD, Allen TM, Birren BW, Henn MR (2012) Highly sensitive and specific detection of rare variants in mixed viral populations from massively parallel sequence data. PLoS Computational Biology 8(3):e1002417.
• Yang X, Charlebois P, Macalalad A, Henn MR, and Zody MC. (2013) V-Phaser 2.0: Variant Inference for Viral Populations. Submitted.

Consequently Mike Zody also spoke at the Workshop on Genomics which I've blogged about regarding NGS experimental design and his publications and presentations are worth a look...

Other Variant Callers:

• PIFs 1.0: Gautier et al., 2013 (site link)   Paper (in revision link)
• SOAPsnp: Li et al., 2009
• Atlas-SNP2: Shen et al., 2010
• SAMtools: a tutorial for how to use SAMtools to variant call
• Another tutorial from ANGUS 2.0 documentation
• The sourceforge site for SAMtools variant calling that my organization blocks...because it's sourceforge (Gah!)
• A wiki from Purdue on calling variants with SAMtools
• FreeBayes: Wiki, Github, googlegroup, Mosaik + FreeBayes pipeline and  Garrison and Marth. 2012
• bcbio-nextgen: documentation and Github
• VarScan: site and Koboldt et al., 2009
• pyroHMMsnp: Zeng, Jiang, and Chen. 2013
• QuRe: Prosperi and Salemi 2012. (for quasispecies reconstruction)

A really nice paper in Nature (disclaimer, it's from 2011 and other programs have come out since) came out talking about SNP/variant callers and provides a referenced table and links to where you can find other information about other programs on SEQwiki.

So you can find a caller to fit your data or NGS experimental design. Just be aware of your programs assumptions and caveats:

• What is the underlying data assumptions of the program
• What was the program 'calibrated on'
• Does it have certain weaknesses to be aware of?
• Has it been compared to other callers? How did it do?
• Does it require you to do backhand springs with your data formatting to even get your data into the caller?
• Are there readily available tutorials or support sites or receptive developers so you don't potentially f*ck it up and if you do you have somewhere to go?
• What are the default parameters? Do they 'make sense' for your data?
• If the defaults don't make sense, do you understand how to change them to fit your data?

Other reading you might consider when evaluating and considering variant calling in your dataset:

• from the GATK forum: Very different SNP calling by GATK and samtools
• Farrer et al., 2013. Using false discovery rates to benchmark SNP-callers in next generation sequencing projects. Scientific Reports 3:1512
• O'Rawe et al., 2013. Low concordance of multiple variant calling pipelines: practical implication for exome and genome sequencing. Genome Medicine 5: 28.
• The trouble with discordant SNP/Indel calling (a blog by NextGenSeek)

Also see the links posted in the previous blog...