The long-term goal of this project is to identify all of the somatic, non-synonymous mutations in the uniqueportion of at least 10 human AML genomes by performing genome-wide, comprehensive resequencing of anumber of tumor samples from carefully selected AML patients; the frequency of these changes will bedetermined in an additional 187 cases of AML. Although the current high cost of whole genomeresequencing is high, we anticipate that costs will continue to fall through the proposed research period.Hence, we will use the data and informatics platforms created by this study to analyze as many AMLgenomes as possible. We have chosen to initially study one of the most common subtypes of AML, FABM1, which has no associated mutations that have been shown to be responsible for the initiation of thisdisease. Samples chosen for analysis have the following characteristics:1. Adequate amounts of bone marrow and skin DMA are available to perform 10X whole genomesequencing coverage, if necessary (i.e. 5 ug of each sample, non-amplified).2. > 70% myeloblasts in the bone marrow sample to assure enrichment of AML cells.3. Two or fewer clonal cytogenetic abnormalities, and complete analysis of tumor and germline DMAsamples on the 500K Affy SNP array and the 2.1 M long oligo CGH array platforms for copy number variants.4. 'Typical' expression signature of M1 AML samples after gene expression profiling analysis.10 FAB M1 samples from our locally banked AML samples meet all criteria, and most have already hadseveral genes resequenced in the initial funding period. Importantly, the detection of mutations in thesesamples by directed PCR and resequencing assures their quality for whole genome resequencing. Usingmaterial from these 10 cases, we propose the following Aims:
Specific Aim 1 : We will use the Solexa massively parallel sequencing platform to perform 30sequencing runs (10X genomic coverage at 1Gb per run) on at least 10 FAB M1 AML samples, andmatched skin DNA samples for the first two AML genomes. Additional AML samples will be analyzed ascost permits.
Specific Aim 2 : We will use the Solexa-based bioinformatics analysis pipeline and our own readmapping strategy to iteratively align the short Solexa reads onto the human reference genome sequence.
Specific Aim 3 : We will develop approaches to identify, validate (in Core D), and annotate all nonsynonymoussomatic mutations in the AML genomes sequenced, following read mapping andidentification of high quality discrepancies. The frequency of these mutations will be evaluated in 187additional AML cases in Core D, and their relevance for pathogenesis and outcomes will be evaluated inother PPG projects.
Showing the most recent 10 out of 122 publications
