Next Generation Read Sequence Assembly Visualization Software

Question

[Neeraj]

Hi

I am working on next generation sequencing Illumina platform. I want to know is there any freely available software for the visualization of the aligned reads including their sequences for Bowtie output (sam or bam format). or software which can read fastq file format.

Edit (md): I was thinking about some requirements for such software:

• Reliably read sam, bam, vcf, and ace, wig files
• Visualization of inserts, deletion, mismatches in the alignments
• Support for visualization of gene models and annotation data in various formats including genbank, embl, gff. For example, if I found my gene of interest I would like to visualize all reads covering a certain part, promotor region or exon.
• Therefore: easy quick navigation in the reads and in the reference annotation data
• ability to view several tracks at the same time
• something like a desktop version of UCSC genome browser + read pilup like in IGV

I think it is possible to do these things with some of the libraries, or partially with some of the desktop tools. But to take a good quick look most tools lack some features.

[Michael]

You notice the contradictions in your question right?

[Michael]

It seems to be a good question though, because a good and easy to use app is missing.

Answer 1

[Gaffa]

See this great blog post reviewing several NGS genome viewers (they all read BAM format).

[Michael]

Thanks, this is great!

Answer 2

[Pablo Pareja]

Hi Neeraj, I'm not sure if it'll exactly be what you're looking for, but Tablet platform is broadly used and freely available.

http://bioinf.scri.ac.uk/tablet/

Cheers,

Pablo

[Michael]

It looks nice but it doesn't want to load my seemingly valid bam file I generated with samtools. That file opened in IGV.

[lh3]

The tablet developers should disable Picard's validation as what IGV is doing.

Answer 3

[Michael]

This question overlaps with What Tools/Libraries Do You Use To Visualize Genomic Feature Data?

Try IGV or IGB for example.

[lh3]

I think this question does not overlap with that one. Displaying NGS alignment data has some special requirements. Although I am very certain that IGV is the best for visualizing NGS variants, I am not 100% sure it is the best for displaying gene annotations or RNA/ChIP-seq.

[Michael]

I think you can use most tools in that thread to build something that fulfills your an my requirements. I'm working with RNA seq, and there it's very important to look at the genome annotation integrated in the browser, while mainly interested in coverage, sometimes one needs to inspect the reads in detail too.

[lh3]

Less than half of tools in that thread display alignments.

Answer 4

[lh3]

IGV is definitely the closest to what you have requested. Highly experienced people at broad who want to visualize SNPs and structural variations have put a lot of stress on IGV, which greatly improves its usability. At least a few months ago, several key features (e.g. multiple alignment tracks, VCF support and remote access) were rarely seen in other viewers. The IGV development group is also very active and responsive.

I have tried other viewers as well. Each of them has some unique interesting features, but they all lack a few key functionalities to be useful for investigating variants and SVs in NGS data. Some of them may be more suitable for RNA/ChIP-seq etc.

On the downside, a few months ago, IGV did not work without a network connection and did not display inserted sequences. I do not know if these have been changed.

EDIT: I am listing several features which I think are key to visualizing variants and SVs. Many of the features may not be needed by visualizing RNA/ChIP-seq.

• Shading based on base quality. Important to investigating SNPs.

• Shading based on mapping quality. Important to SNPs and SVs, but frequently overlooked.

• Displaying unpaired reads. Essential to SVs.

• Multiple alignment tracks. Frequently we want to compare samples.

• Displaying inserted sequences. Important to short INDELs.

• VCF support. VCF is the emerging standard for variants. I know some do not like this format, but there is no better one, at least for now.

• Remote access. This is essential when we want to check a small region of interest from many samples, but do not want to download all the alignment (~20TB for 1000g).

[Michael]

I have started to use IGV myself, mostly out of lack of alternatives, and I don't doubt that the guys at broad are resourceful people. However, if that is the best they could do than there is definitely room for improvement.

[lh3]

IGV is still evolving, which is evident from the long and constantly updated release notes. Almost no software are perfect, due to various tradeoff/balances, different user preferences, constrained time, and perhaps minor design flaws in early days. IGV is not perfect, but in the area I am familiar with, it is by far better than other alternatives, for now.

Answer 5

[Ann]

Integrated Genome Browser (bioviz.org/igb) is also very close to what you requested. It was first developed at Affymetrix and since 2004 is open source and freely available. Several groups contribute to the IGB code base, including mine.

IGV (from Broad) shines in an important use case - visualization of SNPs, as the last post discussed. Depending on users' requests, we'll add features for SNP analysis and visualization that hopefully will build on what IGV can do.

Notable features in IGB:

• Free and open source (sourceforge.net/projects/genoviz/)
• Extensive documentation (see User's Guide)
• Convenient to download and install (uses Java Web Start)
• Shows deletions and mismatches in read alignments (inserts coming soon)
• Click reads to find out read "meta" data, including BAM/SAM header, quality scores, etc
• Can load alignments from the Web or local files (click-drag URLs or use File->Open)
• Support for display of gene models and annotation data in various formats
• Support for data sharing via simple QuickLoad sites (here's an example) or DAS2/GenoPub (requires Tomcat or similar server).
• Highly interactive: Animated zooming, double-click to zoom in on features, click-drag and scrollbars to pan the display, and more
• Supports diverse file formats, e.g., wig, PSL, bed, GFF, and more.
• Allows linking to external databases such as NCBI, TAIR, Phytozome, Affymetrix
• Can view data and genomes from diverse sources, including UCSC and Ensembl
• Edge matching across tracks, useful for comparing boundaries, checking splicing variants
• Bookmarking data scenes, including data sets
• Create annotations and perform manipulations on genome graphs data
• View, copy genome sequence data (useful for designing primers, running BLAST searches)
• Scripting (for power users)
• Easy to create IGB "links," URLs in Web pages that tell IGB to scroll and zoom to a region of interest, load data, etc.

If you have questions or need help, please post to the IGB forum at Sourceforge.

Best wishes,

Ann

[Michael]

The UI of IGB is way too complicated and unstructured. I explain you why, maybe if you are part of the dev-team you could clean that up? Try to perform the following case: you got the newly sequenced genome sequence of B. subtilis strain 666 in a fasta file, the annotation in a Genbank or Embl file. 4 fresh RNA-seq alignments in four bam or sam files (indexed). Try to load all these (the annotation is fresh, it's not on the net), then go to ORF166 and check if some samples indicate a single frameshift mutation at residue 89. But of course everybody is using human or mouse.

Answer 6

[Samuel Lampa]

One of the less known, but recently developed viewers, are the Chipster genome Viewer, available for testing here (second link). (Sourcecode here).

I have not tested it extensively myself, but we were looking into integrating it into Bioclipse because it is designed to be easy to integrate in other software. I heard of some people that IGV is still a bit better though, but as it is a work in progress, it might be worth keeping an eye on ...