Hi everyone,
I am using Nimblegen sequence enrichment and reproducibly getting a lot of short reads on 454 GS Jr, up to 30% sometimes filtered out as short even at cpb ratio 0.1:1, which otherwise gives very good runs apart from wasting the bandwidth to those useless reads. Fragment length distribution in enriched indexed libraries (pools of 6-10 barcoded libs) used to load beads is always beautiful, between 350 and 650 bp and avearging around 450 bp. on Bioanalyzer, as expected from the size distribution of original individual libraries. However, in addtion to a lot of filtered out short reads (I presume adaptor dimers), read length distribution of filter-passed reads consistently shows two peaks, one between 50-100 bp, valley around 150 bp, and another at 450 bp. Nimblegen specs did not offer any explanation. Any similar experience and thoughts on how to overcome this problem?
I am using Nimblegen sequence enrichment and reproducibly getting a lot of short reads on 454 GS Jr, up to 30% sometimes filtered out as short even at cpb ratio 0.1:1, which otherwise gives very good runs apart from wasting the bandwidth to those useless reads. Fragment length distribution in enriched indexed libraries (pools of 6-10 barcoded libs) used to load beads is always beautiful, between 350 and 650 bp and avearging around 450 bp. on Bioanalyzer, as expected from the size distribution of original individual libraries. However, in addtion to a lot of filtered out short reads (I presume adaptor dimers), read length distribution of filter-passed reads consistently shows two peaks, one between 50-100 bp, valley around 150 bp, and another at 450 bp. Nimblegen specs did not offer any explanation. Any similar experience and thoughts on how to overcome this problem?