View Single Post
Old 07-08-2019, 11:02 PM   #230
bartn
Junior Member
 
Location: Europe

Join Date: Jul 2019
Posts: 5
Default randomreads.sh

Hi!

I want to created a (small) RNAseq test set from a small bacterial plasmid. So I used :
Code:
randomreads.sh reads=30000 paired=t metagenome=t  ref=transcripts.fasta out1=test_RNA_1.fq out2=test_RNA_2.fq
(Then I get 30k reads for each pair. No problem, I guess that's because the default is single end)

Then I mapped back the reads with BBmap
Code:
bbmap.sh ref=Rhizobium_Leguminosarum_plasmid_NZ_CP025014.1_transcripts.fasta  in1=Unlock_test_RNA_1.fq in2=Unlock_test_RNA_2.fq covstats=RNA.cov
Results:
Code:
Reads:                               	60000
Mapped reads:                        	54225
Mapped bases:                        	8063372
Ref scaffolds:                       	562
Ref bases:                           	521537

Percent mapped:                      	90.375
Percent proper pairs:                	64.822
Average coverage:                    	15.461
Standard deviation:                    	51.768
Percent scaffolds with any coverage: 	59.96
Percent of reference bases covered:  	45.87
The mapping percentages seem very low to me. Mapping to the genome gives similar results.
I am using randomreads.sh in the wrong way for this?
bartn is offline   Reply With Quote