hts.data_tasks package

Submodules

hts.data_tasks.analysis module

hts.data_tasks.data_normalization module

synopsis

data_normalization implements methods connected to the

readout normalization of a high throughput screening experiment

hts.data_tasks.data_normalization.calculate_local_ssmd(run, **kwargs)

hts.data_tasks.data_normalization.classify_by_cutoff(run, **kwargs)

hts.data_tasks.data_tasks module

hts.data_tasks.data_tasks.perform_task(run, task_name, *args, **kwargs)

hts.data_tasks.qc_detect_data_issues module

synopsis

qc_detect_data_issues implements methods to apply specified thresholds, and declare data that does not

comply with this cut-offs, either in terms of a Run instance with truncated data, or a Data issue file.

hts.data_tasks.qc_detect_data_issues.detect_low_cell_viability(run, control_readout_tag, control_sample_type, controlled_sample_types, data_issue_tag, path, alpha=0.05, *args, **kwargs)

Detect which wells suffer from low cell viability by too strong divergence of a base value. As a practical example, RealTime-Glo measurements are performed for all wells. Sample wells that deviate > x from nominal control wells are considered to have “issues”.

Parameters

• run (run.Run) – A Run instance.

• control_readout_tag (str) – Tag of the control experiment readout (e.g. RealTime-Glo)

• control_sample_type (str) – sample_type used as control.

• controlled_sample_types (list of str) – sample_types that will be evaluated in reference to control_sample_type

• data_issue_tag (str) – Tag of the data issue data that is created.

• path (str) – Path to data issue result file.

• alpha (float) – Significance level for Real-TimeGlo cut-off.

hts.data_tasks.qc_knitr module

synopsis

quality_control implements all methods connected to the

quality control of a high throughput screening experiment. qc_knitr implements knitr specific methods.

hts.data_tasks.qc_knitr.chessboard_pattern()

hts.data_tasks.qc_knitr.compare_plate_replicates(r1=1, r2=2)

hts.data_tasks.qc_knitr.create_report(run, path, methods, force, config_data=None, knit_html=True, resultfile_tag='qc_report', **kwargs)

Run QC & data visualization tasks, and combine the result to a report.

Parameters

• run (run.Run) – Run instance

• path (str) – Path to the resulting qc report file.

• (dict of str (methods) – (dict of str: stuff)): A dictionary connecting an abitrary name of each qc method to a dictionary containing the description (function name, filters, … for the qc method.)

• force (Boolean) – If False, only perform the QC if the result file does not yet exist.

• config_data (list of tuples) – List of tuples used as content for a table in the qc report.

• knit_html (Boolean) –

hts.data_tasks.qc_knitr.dynamics()

hts.data_tasks.qc_knitr.heat_map()

hts.data_tasks.qc_knitr.heat_map_log10_mark_conditionally(condition='y<=-10', color='white')

hts.data_tasks.qc_knitr.heat_map_mark_conditionally(condition='y<=-10', color='white')

hts.data_tasks.qc_knitr.knitr_header_setup(path_knitr_data, plate_names, meta_data=None, original_data_frame='d_all', output='html_document', title='QC Report')

Create knitr markdown file header and setup.

Create knitr markdown file header and setup.

Args:

path_data (str): Path to data file

Returns:

header (str): knitr markdown file header environment (str): knitr markdown environment setter data_loader (str): knitr markdown data loader

hts.data_tasks.qc_knitr.knitr_subset(subset_requirements, original_data_frame='d_all', new_data_frame='d')

Create knitr code to subset the data.

Create knitr code to subset the data.

Args:

subset_requirements (dict): For each requirement, key, values and negated need to be supplied. original_data_frame (str): Name of the original data frame. new_data_frame (str): Name of the new data frame.

Returns:

subset (str): Knittr code subsetting

hts.data_tasks.qc_knitr.kolmogorov_smirnov()

hts.data_tasks.qc_knitr.kolmogorov_smirnov_estimated()

hts.data_tasks.qc_knitr.mean_value_across_plates()

hts.data_tasks.qc_knitr.perform_qc(method_name, *args, **kwargs)

Perform QC method_name with parameter *args and **kwargs, and return result.

hts.data_tasks.qc_knitr.plate_layout()

hts.data_tasks.qc_knitr.replicate_correlation(replicate_defining_column)

hts.data_tasks.qc_knitr.replicate_correlation_robust(replicate_defining_column)

hts.data_tasks.qc_knitr.shapiro_wilk_normality_test()

hts.data_tasks.qc_knitr.smoothed_histogram()

hts.data_tasks.qc_knitr.smoothed_histogram_sample_type()

hts.data_tasks.qc_knitr.ssmd()

hts.data_tasks.qc_knitr.time_course()

hts.data_tasks.qc_knitr.wrap_knitr_chunk(chunk, chunk_name, echo=False, evaluate=True, message=False, warning=False, options='')

Instead of explicitly defining echo, evaluate and so on, it may be advised to implicitly provide all Knitr chunk

options as a string in “options” In the same regard, one could rename “verbosity” to “options”/”knitr_chunk_options”

hts.data_tasks.qc_knitr.z_factor()

hts.data_tasks.qc_knitr.z_prime_factor()

hts.data_tasks.qc_matplotlib module

synopsis

qc_matplotlib implements common plots to visualise HTS data available on the fly.

hts.data_tasks.qc_matplotlib.create_report(*args, **kwargs)

This methods is expected by run.Run . Todo: Needs implementation if Matplotlib reports are required.

hts.data_tasks.qc_matplotlib.heat_map_multiple(run, data_tag, result_file=None, n_plates_max=10, *args, **kwargs)

Create a heat_map for multiple readouts

Create a heat_map for multiple readouts

hts.data_tasks.qc_matplotlib.heat_map_multiple_gaussian_process_model(run, kernel_tag, result_file=None, magnification=5, n_plates_max=10, *args, **kwargs)

Create a heat_map for multiple readouts

Create a heat_map for multiple readouts

hts.data_tasks.qc_matplotlib.heat_map_single(run, data_tag, plate_tag, **kwargs)

Create a heat_map for a single plate

Create a heat_map for a single plate

hts.data_tasks.qc_matplotlib.heat_map_single_gaussian_process_model(run, data_tag_readout, sample_tag, plate_tag, magnification=5, *args, **kwargs)

Create a heat_map for multiple readouts

Create a heat_map for multiple readouts

model_as_gaussian_process(self, data_tag_readout, sample_key,

kernel_type=’m32’, n_max_iterations=1000, plot_kwargs=False,

hts.data_tasks.qc_matplotlib.slice_multiple_gaussian_process_model(run, data_tag_readout, sample_tag, result_file=None, slice=5, n_plates_max=10, *args, **kwargs)

Create a heat_map for multiple readouts

Create a heat_map for multiple readouts

# Currently: using Plotly. # Perhaps, matplotlib.axes._subplots.AxesSubplot can be integrated with Matplotlib’s AxesGrid as above.

hts.data_tasks.qc_matplotlib.slice_single_gaussian_process_model(run, data_tag_readout, sample_tag, plate_tag, slice=5, **kwargs)

Create a heat_map for multiple readouts

Create a heat_map for multiple readouts

# Currently: using Plotly. # Perhaps, matplotlib.axes._subplots.AxesSubplot can be integrated with Matplotlib’s AxesGrid as above.

Module contents