Description:
<jats:title>Abstract</jats:title>
<jats:p>Gene expression profiling has had a significant impact on breast cancer by providing important biological information regarding mechanisms of carcinogenesis, development, and metastasis. In addition, genomic profiling has also resulted in the development and use of new clinical assays for prognostication and prediction. Most genomic profiling has been accomplished using some type of DNA microarray, which is a robust and validated technology where oligonucleotide probes of predetermined sequence are placed/immobilized on known locations on solid surfaces; next, labeled RNAs coming from tumors, cell lines, or animal model tissues are placed onto the array, allowed to hybridize via base complementarity, and then excess probe washed away and intensity scored for each immobilized starting probe. In this way, thousands to millions of probes can be used to interrogate the transcriptome, and in essence, all mRNA genes/transcripts can be interrogated at once.</jats:p>
<jats:p>Technological advances are rapidly occurring and new sequencing-based approaches are gaining use because they offer the ability to gather much more information than conventional DNA microarrays. These sequencing-based approaches are essentially to take all mRNA (or all microRNAs, or some selected fraction of RNA transcripts), make cDNA and sequence everything, thus providing millions of short sequence reads from a given sample/tumor. These sequence reads are then mapped to predetermined transcript libraries of sequences, and the number of reads that map to a given gene/transcript are added up. Finally, computational analyses are used to derive an expression value for each transcript/gene (i.e. RPKM for example). In this way, the number of reads for a given gene are turned into an expression value that is highly reproducible and likely more quantitative than DNA microarrays or possibly qRT-PCR.</jats:p>
<jats:p>There are also many other advantages of the RNA-sequencing approach to the study of gene expression that include 1) mRNA splicing information is obtained and can be used to define transcripts, 2) sequence variation is also obtained so that mutations and/or germline variants can be detected, 3) the presence of fusion genes/proteins can be detected, and 4) the data appears more quantitative and has a larger dynamic range than conventional DNA microarrays. In addition, a number of alternative library preparation protocols can be used. For example, Ribo-depletion or Cot hybridization kinetics allows the removal of highly abundant rRNA transcripts, thus leaving behind all other RNA transcripts and not limiting the detection to PolyA containing transcripts. Another great advantage of sequencing-based approaches is that no prior knowledge of sequence is needed, thus providing a highly objective and complete means of identifying and quantitating transcripts from any species. However, the only limiting features are the number of reads obtained and the costs for a given experiment. In summary, it is likely that, in the long run, sequencing-based approaches may dominate the discovery focused profiling world; however, smaller gene/transcript sets of known composition may still predominate on the many robust gene expression platforms that exist today.</jats:p>
<jats:p>Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr ES7-2.</jats:p>