Low matching from capture libraries

[cmm8cmm8]

Hello,
I have been sequencing NimbleGen Array-captured samples on the SOLiD and seeing strikingly low matching percentages (20-40%). This has occured on human exome captures and mouse captures. Before matching I run a script that finds the NimbleGen adaptor and trims it off the read leaving me with a set of 30bp reads and an untrimmed set of 50bp reads. Both of these sets match poorly, though the 50bp reads are worse. Does anyone have any experience with sequencing captured DNA on the SOLiD? Thanks.

[snetmcom]

how much did you amp up your samples before going on the array?

[cmm8cmm8]

The samples were not amplified prior to hybridization, but the DNA eluted from the arrays was amplified for 27 cycles. Other than possibly introducing additional mutations, why would the amplification cause the matching percentage to drop? We performed matching with up to 5 mismatches, which should still allow the reads to map I would think.

[pmiguel]

Hello,
I have been sequencing NimbleGen Array-captured samples on the SOLiD and seeing strikingly low matching percentages (20-40%). This has occured on human exome captures and mouse captures. Before matching I run a script that finds the NimbleGen adaptor and trims it off the read leaving me with a set of 30bp reads and an untrimmed set of 50bp reads. Both of these sets match poorly, though the 50bp reads are worse. Does anyone have any experience with sequencing captured DNA on the SOLiD? Thanks.

Are you doing the matching in color space (with, for example, Corona-lite)?

--
Phillip