prtecan#
Tecan file has the following structure.
metadata
label 1
label N
Each Labelblock contains: - metadata describing the measurement type and condition; - 96 (float) values, e.g. excitation emission filter-based fluorescence.
A format that can be formalized using the following Backus-Naur form:
TF ::= LABELBLOCK
PRE ::=
LABELBLOCK ::= 'Label: Label'\d+ '\n' ATTRIBUTE+ PLATE ATTRIBUTE+
XXX to be completed…
Object-Oriented Analysis (OOA)#
The program need to be able to:
group Data into 96 dataframes. A single measurement_condition is changed (titrated) over the elements of a group (TecanfilesGroup, LabelblockGroup) and stored into Titration.
export Titration with the properly associated Fitting_function to obtain: fitting_parameters and fitting_quality.
In the case of multi-label file each well contains more than a single measurement under the same measurement_conditions (e.g. ex400nm and ex485nm). Each measurement will be exported into a separate column and Titration could be fitted globally.
Object-Oriented Design (OOD)#
The Metadata class has two private attributes: value and unit. It is used to store metadata values and their corresponding units.
The Labelblock class has several private attributes: lines, path, data, metadata, data_norm, buffer_wells, data_buffersubtracted, data_buffersubtracted_norm, buffer, buffer_sd, buffer_norm, and buffer_norm_sd. It represents a block of labels and its associated data. It has several public methods, including __eq__(), __almost_eq__(), and others.
The Tecanfile class has two private attributes: path and metadata, and a public attribute labelblocks. It represents a Tecan file, and has a one-to-many relationship with Labelblock. It also has a one-to-many relationship with Metadata.
The LabelblocksGroup class has several private attributes: labelblocks, allequal, metadata, data, data_norm, buffer_wells, data_buffersubtracted, and data_buffersubtracted_norm. It represents a group of Labelblock objects and their associated data. It has several public methods.
The TecanfilesGroup class has several private attributes: tecanfiles, labelblocksgroups, and metadata. It represents a group of Tecanfile objects, and has a one-to-many relationship with Tecanfile and LabelblocksGroup.
The Titration class has several private attributes: tecanfiles, conc, additions, buffer_wells, data, and data_nrm. It represents a titration experiment and its associated data. It has several public methods, including load_additions(), and export_data().
The PlateScheme class has several private attributes: file, buffer, crtl, and names. It represents a scheme for a plate, and has several public methods.
The TitrationAnalysis class has several private attributes: scheme, datafit, and heys. It represents an analysis of a titration experiment, and has several public methods, including load_scheme(), fit(), and plot_k(), among others.
Overall, the UML scheme defines several classes that are used to represent a titration experiment, its associated data, and the analysis of the experiment. These classes are organized into groups and have relationships with each other to form a cohesive object-oriented design.
UML scheme#
![class BufferWellsMixin {
+@buffer_wells: list[str]
}
Labelblock .up.|> BufferWellsMixin : Uses >
LabelblocksGroup .up.|> BufferWellsMixin : Uses >
Titration .up.|> BufferWellsMixin : Uses >
class Labelblock{
#_lines: list_of_lines
#path: Path|None
+metadata: dict[str, Metadata]
+data: dict[str, float]
+@data_norm: dict[str, float]
+@buffer_wells: list[str]
+@data_buffersubtracted: dict[str, float]
+@data_buffersubtracted_norm: dict[str, float]
+@buffer: float
+@buffer_sd: float
+@buffer_norm: float
+@buffer_norm_sd: float
__eq__()
__almost_eq__()
}
class LabelblocksGroup{
#labelblocks: list[Labelblock]
#allequal: bool
+metadata: dict[str, Metadata]
+@data: dict[str, list[float]]|None
+@data_norm: dict[str, list[float]]
+@buffer_wells: list[str]
+@data_buffersubtracted: dict[str, list[float]]|None|{}
+@data_buffersubtracted_norm: dict[str, list[float]]|{}
}
class Metadata{
#value: float|int|str
#unit: list[float|int|str]
}
class Tecanfile{
#path: Path
+metadata: dict[str, Metadata]
+labelblocks: list[Labelblock]
}
Tecanfile "1..*" o-- Labelblock
Tecanfile::metadata "*" *-- Metadata
Labelblock::metadata "*" *-- Metadata
LabelblocksGroup::labelblocks "(ordered)" o-- Labelblock
LabelblocksGroup::buffer_wells "0..1" <--> Labelblock::buffer_wells : (same)
class TecanfilesGroup{
#tecanfiles: list[Tecanfile]
+labelblocksgroups: list[LabelblocksGroup]
+metadata: dict
}
TecanfilesGroup "*" o-- LabelblocksGroup
TecanfilesGroup::tecanfiles "1..*" o-- Tecanfile
class Titration{
#tecanfiles: list[Tecanfile]
#conc: ArrayF
#is_ph: bool
#fromlistfile()
+@additions: list[float]
+load_additions(Path)
+@data: list[dict[str, list[float]]|None|{}]
+@data_nrm: list[dict[str, list[float]]|{}]
+@scheme: PlateScheme
+load_scheme(Path)
+export_data()
}
Titration --|> TecanfilesGroup
class PlateScheme{
#file: Path|None
+@buffer: list[str]|[]
+@crtl: list[str]|[]
+@names: dict[str, set[str]]|{}
}
class TitrationAnalysis{
+keys_unk: list[str]
#fromlistfile()
+@fitdata: Sequence[dict[str, list[float]]
+@fitdata_params: dict[str, bool]
+@fitkws: Kwargs
+@results: list[dict[str, FitResult]]
+@result_dfs: list[pd.DataFrame]
+load_scheme(Path)
+fit()
+plot_k()
+plot_all_wells()
+plot_ebar()
+print_fitting()
+plot_buffers()
+export_data()
}
TitrationAnalysis --|> Titration
TitrationAnalysis "0..1" *-- PlateScheme](../_images/plantuml-a93ebd49f5a7793b623430e4e51b0922d96889e7.png)