If you build and copy Bio.Desktop, you can't parse a FASTA File because

Bio.Platform.Helpers.PlatformServices.MaxSequenceSize is set to 0. This leads to a very confusing "sequence is longer than 2GB error when this line of code executes in the FASTA parser.

if ((((long)bufferPosition + line.Length) >= PlatformManager.Services.MaxSequenceSize))
{
    throw new ArgumentOutOfRangeException(
        string.Format(CultureInfo.CurrentUICulture, Properties.Resource.SequenceDataGreaterthan2GB, name));
}

Presumably this is set somehow by nuget or other deployment mechanisms? Do we know how we can make it work for the general build case?

Please add a LICENSE file with the full license text. Additionally it would be great to specify the license (briefly) in the project readme, something along the lines of "Licensed under the Apache 2.0 license".

Bio.Algorithms.Translation.Codons contains the standard genetic code. Should there be a second codon map for mitochondrial genes? AGA and AGG are terminating codons in mitochondria, and UGA codes for tryptophan.

I have noticed that when I try to parse a ClustalW file, it only parses the first set of alignments and ignores all the rest. Is this a common issue and is there a way around it?

This is a wonderful library of functions for bioinformatic analysis. I'm excited to try it out!

Do you have any plans to implement an alignment algorithm similar to STAR, which uses suffix arrays for string searches?

A build on windows or mac with either Xamarin or Visual Studio works fine. Additionally, on Linux using any pre-built CLR package (like our Travis CI builds) also works. However, if installing mono from a tarball using

./configure
make

Then the build fails with an error message indicating that the PCL targets cannot be found.

Are you planning to integrate a VCF parser, or is there another parser I could use for VCF files?

I see there's one in development over on this repository: https://github.com/evolvedmicrobe/Bio.VCF.

Thanks,

AC

In testing pattern searches with BoyerMoore i can't get it to return any matches until the search string starts with a *. This also results in position 0 always being the match location. I'm going to try and do further testing but ever the simple example from the cookbook doesn't behave as i would expect.

Hi,

I am getting following error even in basic code using dotnetbio on windows and visual stduio 2015 community edition.

Code: // Create two sequences; normally you'd have this already.
ISequence dna1 = new Sequence(Alphabets.AmbiguousDNA, "ACTGAAGGATATTA");

Error message
The type initializer for 'Bio.Alphabets' threw an exception.

Thx for your help,
Ambi.

Is it possible to consider an implementation of the Burrows-Wheeler Aligner (BWA), considering it's one of the most popular open-source sequence mapping packages available and is currently only available via a native Linux implementation?

https://github.com/lh3/bwa

There have been occasional Windows ports, but none were official, nor fully functional.

I realize comments aren't documented (and thus probably not supported by) the NCBI docs linked in the parser file, but I feel this is something simple to support and makes the parser a little less strict as comments are simply ignored. Happy to send a PR for the feature if it's something you're interested in

I used the tools before they became a part of OuterCurve. Now you've move them to GitHub.
I tried to compile the libraries to build the tools but I get this error. What type of project do I need to create for nuget to install correctly?

I'm trying to install the .NET BIO libraries, https://github.com/dotnetbio/bio , from Nuget but when I enter the correct commands into the package console I get this error:

PM> Install-Package NETBioCore.PCL

Install-Package : The current environment doesn't have a solution open.

At line:1 char:1

• Install-Package NETBioCore.PCL

• • CategoryInfo : InvalidOperation: (:) [Install-Package], Invalid

  OperationException

  • FullyQualifiedErrorId : NuGetNoActiveSolution,NuGet.PackageManagement.Po

  werShellCmdlets.InstallPackageCommand

I've seen a situation pop up numerous times where the length of the final consensus sequence when calling the following code, changes, if the sequences are ordered from longest to shortest and vice versa, or if the orientation of the sequences changes
Bio.Algorithms.Assembly.OverlapDeNovoAssembler assem = new Bio.Algorithms.Assembly.OverlapDeNovoAssembler();
assem.OverlapAlgorithm.GapOpenCost = -10;
assem.OverlapAlgorithm.GapExtensionCost = -2;
assem.OverlapAlgorithm.SimilarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousDna);
var assembly = assem.Assemble(reads) as Bio.Algorithms.Assembly.OverlapDeNovoAssembly;

compare by assembly.Contigs.First().Consensus.Count
Ive tried to make some simulated data to provide a test case, but can't seem to find one that works.
but I can verify it does this with as little as two sequences

There's a check for GFF parsing that requires version 2. The parser works fine for version 3 (default version for Ensembl now), since the attributes are stored as free text. I'd suggest changing the version check or removing it if the parser also works fine for version 1.

Currently the way to explore the usage of bio.dll is to create a console application that uses bio.dll in some way. This can be made simpler by creating a wrapper powershell module for bio.dll that would allow a new user to try out the various features of the library in an easy way. This would enable the exploration of the library to be much simpler than creating a console application, and encourage new users to try out and create sample/prototype scripts for their end-use.

What do you think about this?

A low-priority issue, but it would be nice to eventually add the three-letter abbreviations for amino acids to the aminoAcidValueMap in ProteinAlphabet.

I've got come code which parses many intervals from a bam file, with calls to the ParseRange() method. The results are correct, but slower than expected.

Reading through the source code (and running the profiler), it looks like the slowdown came from time spent re-parsing the bam header and index in each each call to ParseRange(). Most of my time was spent in BamParser.GetHeader() or BamIndexStorage.Read(). For the particular use case I'm working on, the overhead of re-reading and re-parsing is high. (My bam file has >10000 contigs. I'm parsing several thousand intervals. There are ~100 aligned reads for a typical interval, though some have far higher coverage).

I'd like to speed this up, preferably in a non-hacky way. I've got a change to my copy of the code that does the trick for my use case. The parser caches its I added an optional "useCaching" flag to the ParseRange() method (false by default). If the flag is set to true, the parser re-use the existing SAMAlignmentHeader and BAMIndex objects (if they exist - if they're null we parse as usual). With the current patch it's up to the caller to avoid setting this flag to true when switching between bam files, though that could be an easy sanity-check to add. (And of course, this caching would not be appropriate when parsing a bam file that has been updated on disk in between one ParseRange() call and the next). I can polish that up for a pull request if that sounds like a sane way to proceed and not too much of a corner case to worry about.

I have to revisit an old .NET bio project from a few years back. However, in the meantime I have left the Windows world behind. So I installed MonoDevelop (on Ubuntu 16.04) to see if it would work. I was trying to migrate the old code but it gives problems with simple things that worked fine. For example this:

    Sequence tsequence = new Sequence(Bio.Alphabets.RNA, line1.ToString());

Now throws an exceptions saying "The type initializer for 'Bio.Alphabets' threw an exception."
Apparently I need to "add a reference to a Bio.Platform.Helpers library."
Any ideas?

Today I used dotnetbio to parse a bam file. In rare instances (for a total of 3 reads out of 1 million), BamParser.Parse() returned a null object rather than a SAMAlignedSequence object.

I stepped through in the debugger and I think these lines, in BamParser.cs, are the cause of my issue:
if (alignedSeq.RefEndPos - 1 < start && alignedSeq.RName!="*")
{
return null;
}

Here I'm reading the whole bam, so start == 0. This is a read which is unmapped, but it still has its Pos set to 1 and RName set because its mate was mapped with Pos==1. For the affected read, alignedSeq.RefEndPos is 0. By subtracting 1 from alignedSeq.RefEndPos we get -1, which is less than start == 0, so we return null.

This one-line change fixes the bug, and I believe it's correct in general - I confirmed the unit tests still pass:
if (alignedSeq.RefEndPos - 1 < start && alignedSeq.CIGAR != "*")
{
return null;
}

// [2] Prepare data
string seq = @"GACGCCGCCGCCACCACCGCCACCGCCGCAGCAGAAGCAGCGCACCGCAGGAGGGAAG";
seq.ToString();
Sequence sequence = new Sequence(Alphabets.DNA, seq);

        // [3] Create and configure service handler
        EbiWuBlastHandler blastService = new EbiWuBlastHandler();
       // NCBIBlastHandler blastService = new NCBIBlastHandler();

        ConfigParameters configParams = new ConfigParameters();
        configParams.UseBrowserProxy = true;
        blastService.Configuration = configParams;

        // [4] Define query.
        BlastParameters searchParams = new BlastParameters();
        searchParams.Add("Program", "blastn");

        searchParams.Add("Database", "em_rel");
        searchParams.Add("Expect", "1e-10");
        //searchParams.Add("Email", "YourAddress@YourInstitution");

        // [5] create and submit request
        string jobID;
        try
        {
            jobID = blastService.SubmitRequest(sequence, searchParams);
        }
        catch (Exception ex)
        {
            Console.WriteLine("Service is not available.");
            Console.WriteLine(ex);
            return;
        }

        // [6] Wait for Ready status
        ServiceRequestInformation info = blastService.GetRequestStatus(jobID);

        if (info.Status != ServiceRequestStatus.Waiting
               && info.Status != ServiceRequestStatus.Ready)
        {
            Console.WriteLine("Service is not ready or waiting.");
            return;
        }

        int maxAttempts = 10;
        int attempt = 1;
        while (attempt <= maxAttempts
                && info.Status != ServiceRequestStatus.Error
                && info.Status != ServiceRequestStatus.Ready)
        {
            ++attempt;
            info = blastService.GetRequestStatus(jobID);
            Thread.Sleep(
                info.Status == ServiceRequestStatus.Waiting
                || info.Status == ServiceRequestStatus.Queued
                ? 20000 * attempt
                : 0);
        }

        // [7] Get results
        IList<BlastResult> results2 =
              blastService.FetchResultsSync(jobID, searchParams) as List<BlastResult>;

I'm running this code mentioned in MBF programming guide but it is giving me an error that sequence type parameter is required: Service not available

I want to build this project in Visual studio that's why i need the server information for .Net Bio project.

https://bugzilla.xamarin.com/show_bug.cgi?id=20630