Powered by
,
,
and
,
,
and
Information included: one number metrics (columns 2-15),
visualization reports of metrics (column 1) and
submitting information (columns 16-20).
Once another pipeline submitted, its results will
be automatically updated when estimation finishes.
*More details or customized evaluation refer to our
paper
and Bioconductor package
rnaseqcomp.
| nickname | SD low | SD med | SD high | NE k=1 | NN k=1 | NE k=3 | NN k=3 | TxDiff low | TxDiff med | TxDiff high | deFC low | deFC med | deFC high | pAUC | Run time | Run mem | Pipeline full name | logs | Submit user |
|---|
| nickname | SD low | SD med | SD high | NE k=1 | NN k=1 | NE k=3 | NN k=3 | TxDiff low | TxDiff med | TxDiff high | deFC low | deFC med | deFC high | pAUC | Run time | Run mem | Pipeline full name | logs | Submit user |
|---|
| nickname | SD low | SD med | SD high | NE k=1 | NN k=1 | NE k=3 | NN k=3 | TxDiff low | TxDiff med | TxDiff high | deFC low | deFC med | deFC high | pAUC | Run time | Run mem | Pipeline full name | logs | Submit user |
|---|
| nickname | SD low | SD med | SD high | NE k=1 | NN k=1 | NE k=3 | NN k=3 | TxDiff low | TxDiff med | TxDiff high | deFC low | deFC med | deFC high | pAUC | Run time | Run mem | Pipeline full name | logs | Submit user |
|---|
| nickname | SD low | SD med | SD high | NE k=1 | NN k=1 | NE k=3 | NN k=3 | TxDiff low | TxDiff med | TxDiff high | deFC low | deFC med | deFC high | pAUC | Run time | Run mem | Pipeline full name | logs | Submit user |
|---|
Median standard deviation of quantifications for low abundunce transcripts. Average based on two cell lines is shown.
Similar to "SD low" except for medium abundance transcripts.
Similar to "SD low" except for high abundance transcripts.
In any two compared replicates, proportion of transcripts that expressed in one
but the other, when setting all detrended logsignals below 1 to 0.
Results from all two replicate comparisons are averaged.
Then, average based on two cell lines is shown.
In any two compared replicates, proportion of transcripts that expressed in neither replciates,
when setting all detrended logsignals below 1 to 0.
Results from all two replicate comparisons are averaged.
Then, average based on two cell lines is shown.
Similar to "NE k=1" except that all detrended logsignals below 3 are set to 0.
Similar to "NN k=1" except that all detrended logsignals below 3 are set to 0.
For low abundunce genes that have only 2 annotated transcripts, the absolute proportion
differences for 1 of the two transcripts
between any two compared replicate samples.
Mean of results from all two replicate comparisons is first calculated.
Then, average based on two cell lines is shown.
Similar to "TxDiff low" except for medium abundance transcripts.
Similar to "TxDiff low" except for high abundance transcripts.
For low abundunce pre-determined differential expression transcripts, the median of their estimated log2 fold changes between two cell lines based on average signals across replicates. Note here: log2 fold changes are set to positive if they agree with the sign of pre-determined fold changes.
Similar to "deFC low" except for medium abundance transcripts.
Similar to "deFC low" except for high abundance transcripts.
Differential expression is first estimated between each replicate in one cell line and each replicate in the other. Fold changes are used to make ROC curves with comparison to the TRUE differentials defined by microarray (ENCODE dataset) or pre-determined (Simulation dataset). All the ROC curves are further averaged based on different thresholds of fold changes. Partial AUC with false positive rate ranges from 0 to 0.2 is then standardized as pAUC.
Note: Time and memory costs are based on quantification process and
alignment process (if used) of 1st sample in each dataset.
Estimation is based on 8 parallel cores for SIMULATION datasets
and 12 cores for ENCODE dataset.
|
Powered by
,
,
and
|
||
| This website designed and maintained by Mingxiang Teng and Rafael A. Irizarry | ||