using spike-in funcitonality

[zuta-osa77]

Hello do you have an example of using the -spk or --spikeIn functionality in miRge3.0? In particular, an example of calling and how the bowtie index library looks.

Specifically

a) I have put the bowtie indices for the separate spike-in sequences where the human indices are located as instructed, but annotation.report.csv still reports 0 reads for the Spike-In category?

b) can > 1 spike-in sequences be used?

thanks

[arunhpatil]

@zuta-osa77 ,

Yes, we have bowtie index data available for spike-in in the libraries (described below), and providing the parameter -spk will map the reads to spike-in data along with other databases. If you see 0 reads in the annotation reports suggests that the input FASTQ files may have different spiked-in reads or may not contain any.

If you do list directory command, the index files for spike-in is part of other listing (only showing the spike-in index files below) :
ls miRge3_Lib/human/index.Libs/

human_spike-in.1.ebwt      
human_spike-in.2.ebwt      
human_spike-in.3.ebwt      
human_spike-in.4.ebwt      
human_spike-in.rev.1.ebwt  
human_spike-in.rev.2.ebwt

Now, if you want to see the contents of it, you can use the following command (only showing the first four lines of output here):
bowtie-inspect miRge3_Lib/human/index.Libs/human_spike-in | head

>Hafner_Cal01
GTCCCACTCCGTAGATCTGTTC
>Hafner_Cal02
GATGTAACGAGTTGGAATGCAA

You may use less instead of head to browse more.

An example usage would be:
miRge3.0 -s SRR772403.fastq -lib /mnt/d/Halushka_lab/Arun/datasets/miRge3_Lib -on human -db miRGeneDB -o output_dir -a illumina -spk

If you think you need to replace spike-in data already indexed here then, you may follow the commands below:
NOTE: If you want to append to the existing library, then the instructions are provided in the next section: append*.

1. rm -rf human_spike-in.* (remove the existing files with prefix human_spike-in)
2. bowtie-build [options]* <reference_in> <ebwt_outfile_base> (Command to create new index files)
   example: bowtie-build --threads 6 my_spikein.fasta human_spike-in (command to create, example)
   my_spikein.fasta: should contain spike-in sequences in FASTA format.

This will create the bowtie ebwt files with name prefix human_spike-in and it is mandatory to retain this name format in the library directory.

append* :
To append to an existing spike-in library:

1. bowtie-inspect miRge3_Lib/human/index.Libs/human_spike-in > my_spikein.fasta - This command will create a copy of the sequence from ebwt file to fasta file named as my_spikein.fasta
2. append your own sequences at the bottom of this file , you could cat inputFile.fasta >> my_spikein.fasta or if you are familiar vim editior you could open it and append at the bottom or simply open in a notepad editor (gedit like editor) and append at the bottom. Please make sure, the file has to be in FASTA format.
3. Index the files back to human_spike-in using the bowtie-build command shown earlier.

I hope this is clear, let me know if you need any additional information.

Thank you,
Arun.

[zuta-osa77]

Thanks Arun, the appending process worked!

By having all the spike-in reads aggregated however we cannot resolve different concentrations of multiple spike-in sequences (eg for RNA isolation control). In this case we would need to generate counts for each spike-in sequence -- I can do this manually with bowtie but I'm wondering if there's a hack within the miRge3.0 space to carry this out.

[arunhpatil]

@zuta-osa77,

I wonder if you have different concentrations of spike-in across conditions, then that should reflect in the mapped.csv file.
see, for example, the column spike-in is null and represents 16 counts for isomiR for sequence TCCCTGATACCCTAACTTGTGA

Sequence,annotFlag,exact miRNA,hairpin miRNA,mature tRNA,primary tRNA,snoRNA,rRNA,ncrna others,mRNA,isomiR miRNA,`spike-in`,SRR772403
TCCCTGATACCCTAACTTGTGA,1,,,,,,,,,Hsa-Mir-10-P3b_5p,,16

Then you could use this information across conditions and spike-in counts to your advantage.

Thank you,
Arun