So you have two things you would like to accomplish downloading a list of SRA files, and converting those files to Fastq files. You likely want your files in fastq format, so you will need to convert them. Starting at the very beginning, SRA uses their own internal format to compress files. I spend WAY too much time dealing with this database and will share my methods. +SRR1951502.2.2 HISEQ:242:C5KU3ANXX:7:1101:1377:2335 length=100īBBBBFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF NTTTAAGATAGTTATCTGGTTGATCCTGCCAGTAGTCATATGCTTGTCTCAAAGATTAAGCCATGCATGTCTAAGTATAAGCAATTTATACAGTGAAACT GCAATGATTAACTTAGTACCGTTCTACCAAGTACGCCCTTGATTACCTCATTAACGAACAAAAACGCAGATCCCTTCCTAAGGCGGCCGAGACGGCCGCT Looks like reddit murders the formatting, but whatever. after doing a bit of investigation myself, it looks like NCBI moved everything to https, which breaks compatibility with all versions of sratoolkit earlier than 2.8.0 (see for the updated version and news). Your first step should be to run FastQC on them. You now have an output directory filled with fastq files. # add the -X 5 option to make sure things are working as a test (only dumps 5 reads per file)įastq-dump -split-files -gzip -I -O /path/to/output/directory $SAMPLEīoom. # make sure these are the options you want: When you're ready, run the following script (if you're feeling hardcore/pressed for time, you can parallelize this with tools like Snakemake): #!/bin/bashįor SAMPLE in $(cat SRR_Acc_List.txt) do If not, you're about to fry your hard disk. Make sure you've got the SRA Toolkit installed on a Linux cluster somewhere with some serious hard drive space.It's your responsibility to make sure you're downloading the right data here. Download both the Accession List (which samples you want) and RunInfo Table (you only need this later, when figuring out what's what) for the samples you want.Clicking on "All Runs" under "Study" gets me here:.Clicking on the middle one gets me here:.Type in your accession into the search bar here (looks like there were 3 studies matching :.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |