Hey all,
Small RNA libraries were prepared by using illumina TrueSeq small RNA library and NGS was performed using illumnia Solexa technology. Now i have these fastq files from the plasma having small non coding RNAs. I already did some analysis and have results but want to onfirm what i did is right or not, suggestions are always welcomed. At first I actually want to know the total landscape of all small non coding RNA in the samples (control and cases).
I used trimmomatic to remove the adpators and do quality filetring:
java -jar trimmomatic-0.32.jar PE -threads 24 -phred33 input/20160332_ATCACG_R1.fastq.gz input_20160332_ATCACG_R2.fastq.gz trim_output/20160332_ATCACG_R1P.fastq.gz trim_output/20160332_ATCACG_R1U.fastq.gz trim_output/20160332_ATCACG_R2P.fastq.gz trim_output/20160332_ATCACG_R2U.fastq.gz ILLUMINACLIP:TrueAll.fa:0:30:10:4:true SLIDINGWINDOW:4:15 LEADING:3 TRAILING:3 MINLEN:25
Want to know the parameters i used are right? 1) in trimmomatic package, a folder named adapter has teh TruSeq sequences, i concatenated all teh files in the folder (SE, PE, NexteraPE) and made one TrueAll.fa file and used that for adaptor trimming? 2) mismatehes=0 3) minimun length =25.
After trimming the total sequences dropped from 3514467 to 213003, and sequence length changed from 101 to 18-101. Did i loose alot of reads after trimming?
Then, after quality control, i used Tophat2 (default setting) for alignment using UCSC hg19 genome.
Is the approch OK? i read somwhere that GENCODE annotation is much better if we have to work on the small non coding features.
Please, give me suggestions about the right way of analysing my plasma small non coding RNA. Thanks!!!!!!!
I found http://miarmaseq.cbbio.es/ really an amazing tool for circular sRNA-Seq, mRNA-Seq and totally sRNA-Seq data. but unfortunately I was not able to install a prerequisite. I hope you could run this comprehensive pipeline on your data.
I must admit that I also first thought it was about circular RNA, but it's actually circulatory, i.e. circulating in the blood/plasma ;)
@WouterDeCoster I really liked this tool and am regret fail to run that :)
Thanks Angel. i will try using this pipeline..
Well for the samples i have i want to see all possible small non coding RNA present.
But i really want to know about the trimming i did, is that right or what approach should i use.
I only have experience in working with Cutadapt and BBDuk in adapter removal and found both amazing in convenience and flexibility in changing parameters. I have not used trimmomatic yet.
miARma-Seq pipeline does adapter removal and quality filtering automatic and only needs fastq files as input. even if you are working with homo sapiens this tool has pre-built genome index for download.
If you could take some time trying that finally it would give you a read count file of non-coding sRNA corresponding to your samples.
however I recently performed sRNA-Seq and set minimum length to 15 and removed illumina universal small RNA seq 3' adapter** existed in trimmomatic adapter file.
For the first point, I do not know how Trimmomatic approaches files with loads of adapter sequences. I rather use the trimm galore approach of just giving the shared sequence for Illumina adapters for the software to remove.
Also, about the minimum length, it really varies depending on what you are trying to find. When I work with miRNAs, usually anything bellow 17-18 bp makes it very difficult to classify. But it depends what are you looking for and how did you prepare the libraries (maybe if you are interested on a certain fraction of the small RNAs, you can overlook the rest?)