Transcript concentrations in cancer cell lines and cervix tumors

We determined the transcript concentration of 10157 genes and ESTs in twelve cervical tumors and a pool of ten cancer cell lines. Following this link, a table is found where we listed estimated concentrations for each gene and for each tumor, equipped with its 95% credibility interval.

Similarly, following this link, for the cancer cell lines, we reported estimated concentrations and credibility intervals for each gene. We also included the mean concentrations of the cervix tumors.

All concentrations are given in number of transcripts per microgram of total RNA.

The results above are obtained from data where no backgound correction is included. Results for data where background correction has been included is found here: Tvelwe cervix tumors and Mean patient and Reference.

TransCount: Genome-wide estimation of transcript concentrations from spotted cDNA microarray data.

TransCount is a program for estimating absolute transcript concentrations and other experimental parameters in a microarray experiment. The model, the estimation method implemented and the kind of data sets assumed are described in detail in the submitted paper "Genome-wide estimation of transcript concentrations from spotted cDNA microarray data." by Arnoldo Frigessi, Mark A. van de Wiel, Marit Holden, Debbie H. Svendsrud, Ingrid K. Glad and Heidi Lyng. An additional technical report is available here. All data sets used in the paper and the corresponding MCMC traces are available below, ready for further statistical analysis.

Software for obtaining MCMC traces and absolute transcript concentrations will be publicly available here, as a userfriendly BASE plugin. BASE is a comprehensive free web-based database solution for the massive amounts of data generated by microarray analysis. The plugin, called Transcount, is also available here together with a demo data set. A short description of how to use BASE and how to run the Transcount plugin from BASE, is given on that web page.

For generating the MCMC traces available below, the two C++-programs transCount and transCountVal were used. A description of these programs and binary code for them are available here.

Spike experiments - Figure 2 (two dye-swaps)

Data sets as formatted input files to the transCountVal program

First dye-swap and second dye-swap.

Output files, results: MCMC traces from the transCountVal program

Results after 5 000 000 iterations, after a burn-in of 500 000 iterations, and thinning 1000 iterations (storing every 1000 iteration), using the transCountVal program: First dye-swap and second dye-swap.

Cancer cell line and cervix cancer experiment - Figures 3-7

Data sets as formatted input files to the transCount program

Cancer cell line and cervix cancer and Spikes

Output files, results: MCMC traces from the transCount program

Results after 250 000 iterations, including a burn-in of 50 000 iterations, and thinning 100 iterations, using the transCount program: Cancer cell line and cervix cancer (info file, last file, res file, geneRes file, Ktilde0Res file, Ktilde1Res file, Ktilde2Res file, Ktilde3Res file, Ktilde4Res file, Ktilde5Res file, Ktilde6Res file, Ktilde7Res file, Ktilde8Res file, Ktilde9Res file, Ktilde10Res file, Ktilde11Res file, Ktilde12Res file) and Spikes (info file, last file, res file, geneRes file, Ktilde0Res file, Ktilde1Res file)