Negative controls and index hopping Illumina HiSeq - Metagenomics
Hi there,
We consistently conduct multiplexed shotgun metagenomic sequencing (human stool samples) using an Illumina PCR-free protocol. We were previously sequencing on the HiSeq 2500 but have recently begun sequencing on the HiSeq X.
We usually multiplex 48 samples per pool/lane with one negative control per pool (negative control starting from extraction right through to sequencing). From this we usually obtain 4-8 million reads per sample. Previously on the HiSeq 2500 we obtained 10-70,000 reads in the negative controls. Since switching to the HiSeq X, we have obtained up to 600,000 reads in our negative controls.
We do not believe that this is true contamination in the negative controls as they record no DNA concentration following DNA extraction (picogreen), no tapestation signal and negligible qPCR values using the KAPA library quantification kit following library preparation (~2pM vs 10nM for actual samples). Therefore, prior to sequencing, there does not seem to be any relevant quantity of DNA/library in the negative controls.
We therefore think that this may be a sequencing/bioinformatic issue. We are aware of potential index hopping (we are using single index adapters but hopefully switching to dual index as a result of this), however we are unsure if index hopping could contribute to this number of reads in a negative control?
Any insight into this issue would be very much appreciated as we are unsure what to do with our sequencing results with so many reads in our 'negative' controls!
Thanks!
We consistently conduct multiplexed shotgun metagenomic sequencing (human stool samples) using an Illumina PCR-free protocol. We were previously sequencing on the HiSeq 2500 but have recently begun sequencing on the HiSeq X.
We usually multiplex 48 samples per pool/lane with one negative control per pool (negative control starting from extraction right through to sequencing). From this we usually obtain 4-8 million reads per sample. Previously on the HiSeq 2500 we obtained 10-70,000 reads in the negative controls. Since switching to the HiSeq X, we have obtained up to 600,000 reads in our negative controls.
We do not believe that this is true contamination in the negative controls as they record no DNA concentration following DNA extraction (picogreen), no tapestation signal and negligible qPCR values using the KAPA library quantification kit following library preparation (~2pM vs 10nM for actual samples). Therefore, prior to sequencing, there does not seem to be any relevant quantity of DNA/library in the negative controls.
We therefore think that this may be a sequencing/bioinformatic issue. We are aware of potential index hopping (we are using single index adapters but hopefully switching to dual index as a result of this), however we are unsure if index hopping could contribute to this number of reads in a negative control?
Any insight into this issue would be very much appreciated as we are unsure what to do with our sequencing results with so many reads in our 'negative' controls!
Thanks!