As of April 20, two new machines have been added to campusrocks, each with 48 cores and 192 GB of RAM. These machines appear to be good candidates for running the more memory intensive assemblers (SOAPdenovo, ALLPATHS, and ABySS).

We noticed a discrepancy in the total k-mer counts found using the histogram and the total number of k-mers from the jellyfish stats command. It turned out that there were k-mer counts the were much higher than the 500 and later 5,000 cut off we applied using the jellyfish histo command. For the first illumina run the maximum count found for a 22-mer was 23,661,844, and in run 2 the maximum count for a 22-mer was 6,090,098. The k-mers with extremely high counts are likely the result of contamination or an artifact from the library preparation. We need to examine these k-mers to find where they came from. One possible solution is to run SeqPrep before counting k-mers to remove adapter sequences from the reads.

We will do paper presentations on Monday and Wednesday next week.