cmuts norm¶
Purpose¶
Experiment-independent normalization of reactivity rates.
Requires¶
One or more sets of reactivity rates, in
cmuts-compatible HDF5 files
Usage¶
Every input given to one run shares a single norm, and each is written to an output of its own. -o is repeated once per input and paired with them in order.
cmuts norm -o apo-normalized.h5 -o holo-normalized.h5 apo.h5 holo.h5
Running the inputs separately gives each its own norm instead.
cmuts norm -o apo-normalized.h5 apo.h5
cmuts norm -o holo-normalized.h5 holo.h5
This is only suggested if the two experiments had significantly different conditions which are not directly comparable.
Normalization Schemes¶
All normalization schemes except value compute the norm using the aggregated mismatch, insertion, and deletion rates. Only the positions whose coverage exceeds --min-coverage contribute to it.
The default ubr scheme computes the norm as the 90th percentile of the aggregate rate. The outlier scheme drops the highest 2% of the aggregate rate as outliers and averages what lies between there and the highest 10%. The value scheme uses the norm provided in --value.
A norm that comes out as zero, negative, or undefined is not applied and is recorded as NaN.
Output¶
Dataset |
Source |
|---|---|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Divided by the norm. |
|
Estimated from the input per the specified scheme. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
|
Copied from the input. |
CLI Options¶
Arguments¶
Argument |
Description |
|---|---|
|
the reactivities to normalize |
Input and output¶
Option |
Description |
|---|---|
|
write one input’s results to this file; repeat once per input (required) |
|
replace the output files if they already exist |
Normalization¶
Option |
Description |
|---|---|
|
where the norm comes from (ubr, outlier, value; default ubr) |
|
divide by this norm, in place of computing one (with |
|
coverage a position needs before its rate sets the norm (with |
Information¶
Option |
Description |
|---|---|
|
show this help and exit |
|
show the version and exit |
Advanced¶
Accepted, and left out of --help.
Option |
Description |
|---|---|
|
describe every argument as JSON and exit |
|
describe the output format as JSON and exit |