| Name | hts-tools JSON |
| Version |
0.0.2.post0
JSON |
| download |
| home_page | |
| Summary | Parsing and analysing platereader absorbance and fluorescence data. |
| upload_time | 2023-11-16 14:25:29 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | >=3.8 |
| license | MIT License Copyright (c) [year] [fullname] Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. |
| keywords |
science
assay
platereader
analysis
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# 🏮 hts-tools
Parsing and analysing platereader absorbance and fluorescence data.
- [Installation](#installation)
- [Command-line usage](#command-line-usage)
- [First pipeline example](#first-pipeline-example)
- [Parsing platereader exports](#parsing-platereader-exports)
- [Adding experimental conditions](#adding-experimental-conditions)
- [Normalization within batches](#normalization-within-batches)
- [Plotting dose responses](#plotting-dose-response)
- [Second pipeline example](#second-pipeline-example)
- [Converting plate-shaped data to columns](#converting-plate-shaped-data-to-columns)
- [Statistcal testing](#statistical-testing)
- [Other commands](#other-commands)
- [Python API](#python-api)
- [Documentation](#documentation)
## Installation
### The easy way
Install the pre-compiled version from PyPI:
```bash
pip install hts-tools
```
### From source
Clone the repository, then `cd` into it. Then run:
```bash
pip install -e .
```
## Command-line usage
**hts-tools** provides command-line utlities to analyse and plot data
from platereaders, starting with the _raw_ exported data with no
manual conversion or copy-pasting needed. The tools complete specific tasks which
can be easily composed into analysis pipelines, because the TSV table output goes to
`stdout` by default so they can be piped from one tool to another.
To get a list of commands (tools), do
```bash
hts --help
```
And to get help for a specific command, do
```bash
hts <command> --help
```
For the Python API, [see below](#python-api).
## First pipeline example
This command takes several exported Excel files (matching the pattern `plate-?.xlsx`) from a Biotek platereader,
adds annotations on experimental conditions, and normalizes the data based on positive and negative controls.
Finally, dose-response curves are plotted.
```bash
hts parse plate-?.xlsx --data-shape row --vendor Biotek \
| hts join --right layout.xlsx \
| hts normalize --control compound_name \
--positive RIF --negative DMSO \
--grouping plate_id --method npg \
| hts plot-dose -x concentration --facets guide_name --color compound_name \
--output plt-test
```
### Parsing platereader exports
The command `hts parse` converts Excel or CSV or TSV files exported from platereader software from a specified vendor into a
uniform columnar table format used by all downstream hts-tools. The `data-shape` option indicates whether the export was in plate
format or row-wise table format. It should usually be the first command in a pipeline.
`hts parse` produces a table with at least the following columns (with example entries):
| row_id | column_id | well_id | plate_id | data_filename | data_sheet | measured_abs_ch1 | abs_ch1_wavelength |
| ------- | --------- | -------- | -------- | ------------- | ---------------- | ---------------- | ------------------ |
| A | 1 | A01 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.366 | 600nm |
| A | 2 | A02 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.402 | 600nm |
If only fluorescence was measured, the last two columns would be called `measured_fluor_ch1` and `fluor_ch1_wavelength`. If there were multiple
measurements, then they will appear as additional columns with increasing `ch` (channel) numbers, for example, `measured_fluor_ch2`, `fluor_ch2_wavelength`.
There will also be additional columns using information scraped from the input file. These will usually have headings starting with `meta_`,
such as `meta_protocol_file_path`, `meta_date`, `meta_time`, and `meta_reader_serial_number`.
### Adding experimental conditions
The output from `hts` parse is passed to `hts join`, which combines two tables based on values in shared columns (known as a
database join). For example, if `layout.xlsx` contains a sheet with this table:
| column_id | plate_id | compound_name |
| --------- | -------- | ------------- |
| 2 | Plate 6 | trimethoprim |
| 1 | Plate 6 | moxifloxacin |
then `hts parse plate-?.xlsx --data-shape row | hts join --right layout.xlsx` will result in
| row_id | column_id | well_id | plate_id | data_filename | data_sheet | measured_abs_ch1 | abs_ch1_wavelength | compound_name |
| ------- | --------- | -------- | -------- | ------------- | ---------------- | ---------------- | ------------------ | ------------- |
| A | 1 | A01 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.366 | 600nm | moxifloxacin |
| A | 2 | A02 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.402 | 600nm | trimethoprim |
Since `column_id` and `plate_id` are the only column headings in common, the entries of these columns are used to match the rows of the
tables. So where `column_id = 2`, `compound_name = trimethoprim` will be added.
If you join an Excel XLSX file containing multiple sheets, these will each be joined in order. In this way, you can add experimental
conditions easily by, for example, first joining the conditions that vary by plate and row (such as compound), then by column
(such as concentration). This approach is very flexible, and you can join on any number of columns and add any new ones you like as
long as the column headings aren't repeated.
### Normalization within batches
The `hts normalize` command normalizes raw measured data based on controls. This can add power to
downstream statsical analyses by mitigating batch-to-batch variation.
`hts normalize` adds new columns for each measured column. These columns start with `calc_` and end with `_norm.{method}`, for example `calc_abs_ch1_norm.npg` and `calc_abs_ch2_norm.npg`. Two methods are offered which
optionally normalize within within groups, such as batches or plates.
#### Normalized proportion of growth (NPG)
This method scales raw data to be between $0$ and $1$ based on positive _and_ negative
controls, optionally within groups (or batches) of measurements. In the example above,
the positive and negative controls are defined as `RIF` and `DMSO`, and should be found in the column `compound_name` (which may have been added by `hts join`).
The positive and negative controls are averaged within each value in the `--grouping` column. In
the example above, they will be averaged for each `plate_id`, and these will be used to normalize
the measured values of that `plate_id` according to:
$$s = \frac{m - \mu_p}{\mu_n - \mu_p}$$
where $s$ is the normalized value, $m$ is the measured value, and $\mu_p$ and $\mu_n$ are the mean positive and negative controls.
#### Proportion of negative (PON) [default]
This method scales raw data relative negative controls _only_, optionally within groups (or batches) of measurements.
The negative controls are averaged within each value in the `--grouping` column. In
the example above, they would be averaged for each `plate_id`, and these will be used to normalize
the measured values of that `plate_id` according to:
$$s = \frac{m}{\mu_n}$$
where $s$ is the normalized value, $m$ is the measured value, and $\mu_n$ is the mean negative control.
### Plotting dose response
`hts plot-dose` is a very flexible command which takes the columnar tables as input and plots the data in almost any breakdown using a color-blind
palette. The required `-x` option indicates which column to use as the x-axis (usually concentration). The y-axis will be values in all the measured
and calculated columns (`hts plot-dr` plots them all automatically in seaparte files). The other options allow splitting the plots by file, facet
(panel) and color according to the values in columns.
The example above, `hts plot-dose -x concentration --facets guide_name --color compound_name`, will produce plots like this:
<img src="docs/source/_static/plt-test_calc_abs_ch1_norm_.png" alt="" width="600">
The panels each value with the same `guide_name` is in a `facet` (panel), and the lines are colored by `compound_name`.
## Second pipeline example
Here is another example showing the sequential use of `hts join` to join two tables of experimental data, and two other commands: `hts pivot` and
`hts summarize`.
```bash
hts pivot compounds.xlsx \
--name compound_name
hts parse plate-*.txt --data-shape plate \
| hts join --right sample-sheet.csv \
| hts join --right pivoted-compounds.tsv \
| hts normalize --control compound_name --positive RIF --negative DMSO --grouping strain_name plate_id \
| hts summarize --control compound_name --positive RIF --negative DMSO --grouping strain_name compound_name --plot summary \
> summary.tsv
```
These commands are explained below.
### Converting plate shaped data to columns
Sometimes you will want to use data, such as plate layouts, which are in a plate-shaped layout instead of a column format. For example:
<img src="docs/source/_static/plate-screenshot.png" alt="" width="800">
You can convert this to column format using `hts pivot`, which produces a table in the following format:
| row_id | column_id | compound_name | well_id | plate_id | filename |
| ------ | --------- | ------------- | ------- | ------------------------ | ------------------------ |
| C | 2 | RIF | C02 | Plate 7 | compounds.xlsx |
| D | 2 | LYSINE | D02 | Plate 7 | compounds.xlsx |
| E | 2 | DMSO | E02 | Plate 7 | compounds.xlsx |
| F | 2 | RIF | F02 | Plate 7 | compounds.xlsx |
It is assumed that there is one plate per sheet for Excel files, and one plate per file for TSV and CSV files. The
plate name is taken from the sheet name (Excel) or filename (other formats).
You can prepend the names of the `plate_id` and `filename` columns with the `-x` option. for example,
`hts pivot compounds.xlsx -x compound_source --name compound_name` would have columns `compound_source_plate_id`
`compound_source_filename`. This is helpful when usign `hts join` later where the plate and filename
columns would otherwise be shared but have different meanings and values, and you don't want to accidentally
join on them.
### Statistical testing
Groups of values (such as replicates) can be compared against a negative control for statistical testing using
`hts summarize`. The `--grouping` option indicates the columns whose values together indicate values which are
replicates of a particular condition of interest. For example, `--grouping strain_name compound_name` would
indicate that values which have the same `strain_name` and `compound_name` are replicates.
Statistical tests compare to the `--negative` values, and use all measured and normalized (`calc_*_norm`) columns.
Currently, the Student's t-test and Mann-whitney U-test are implemented. The t-test is best suited to Normal-distributed
data while the MWU is better for other distributions which might not have a nice bell curve distribution.
Although `hts summarize` calcualtes both tests simultaneously, it's not a good idea to look for "significant" $p$-values
in both. This is called p-hacking, and leads to false positives. Instead, decide which test is most appropriate for your
data and stick with that one.
This command also calculates other summary statistics such as between-replicate mean, variance, and SSMD. If a filename
prefix is provided to the `--plot` option, then volcano and flashlight plots are produced, which may be useful for identifying
hits of high throughput screens.
<img src="docs/source/_static/summary-counter_fluor_ch1_norm_volcano.png" alt="" width="600">
<img src="docs/source/_static/summary-counter_fluor_ch1_norm_flashlight.png" alt="" width="600">
## Other commands
There are several other commands from **hts-tools** which take as input the output from `hts parse`, joined
to an experimental data table (`layout.xlsx` in these examples).
- `hts qc`
Do quality cpntrol checks by calculating mean, standard deviation, Z\'-factor, and SSMD, and plotting them.
```bash
hts parse plate-?.xlsx --data-shape row \
| hts join --right layout.xlsx \
| hts qc --control compound_name --positive RIF --negative DMSO --grouping strain_name plate_id --plot qc-plot \
> qc.tsv
```
<img src="docs/source/_static/qc-plot_fluor_ch1_mean_sd.png" alt="" width="600">
<img src="docs/source/_static/qc-plot_fluor_ch1_zprime.png" alt="" width="600">
- `hts plot-hm`
Plot heatmaps of signal intensity arranged by plate well. This can be useful to identify unwanted within-plate variability.
```bash
hts parse plate-?.xlsx --data-shape row \
| hts join --right layout.xlsx \
| hts plot-hm --grouping strain_name sample_id plate_id --output hm
```
Here, `--grouping` identifies the columns which indicate values coming from the same plate. One file is produced per
measured and normalized (`calc_*_norm`) column.
<img src="docs/source/_static/hm_calc_fluor_ch1_norm_heatmap.png" alt="" width="600">
- `hts plot-rep`
Plot two replicates against each other for each condition.
```bash
hts parse plate-?.xlsx --data-shape row \
| hts join --right layout.xlsx \
| hts plot-rep --control compound_name --positive RIF --negative DMSO --grouping strain_name compound_name --output rep
```
Here, `--grouping` identifies the unique conditions within which values are treated as replicates. The positives and negatives
are plotted as different colors. One file is produced per measured and normalized (`calc_*_norm`) column.
<img src="docs/source/_static/rep_measured_fluor_ch1_replicates.png" alt="" width="600">
In the plots, the left column is on a linear-linear scale and the right column is on a log-log scale. There is one row
of plots per wavelength set in the dataset.
- `hts plot-hist`
Plot histograms of the data values.
```bash
hts parse plate-?.xlsx --data-shape row \
| hts join --right layout.xlsx \
| hts plot-hist --control compound_name --positive RIF --negative DMSO --output hist
```
The positives and negatives are plotted as different colors. One file is produced per measured and normalized (`calc_*_norm`) column.
<img src="docs/source/_static/hist_measured_fluor_ch1_histogram.png" alt="" width="600">
In the plots, the left two columns are on a linear-linear scale and the right two columns are on a log-log scale. There is one row
of plots per wavelength set in the dataset.
## Python API
**hts-tools** can be imported into Python to help make custom analyses.
```python
>>> import htstools as hts
```
You can read raw exports from platereader software into a columnar Pandas dataframe.
```python
>>> hts.from_platereader("plates.xlsx", shape="plate", vendor="Biotek")
```
Once in the columnar format, you can annotate experimental conditions.
```python
>>> import pandas as pd
>>> a = pd.DataFrame(dict(column=['A', 'B', 'A', 'B'],
... abs=[.1, .2, .23, .11]))
>>> a
column abs
0 A 0.10
1 B 0.20
2 A 0.23
3 B 0.11
>>> b = pd.DataFrame(dict(column=['B', 'A'],
... drug=['TMP', 'RIF']))
>>> b
column drug
0 B TMP
1 A RIF
>>> shared_cols, data = join(a, b)
>>> shared_cols
('column',)
>>> data
column abs drug
0 A 0.10 RIF
1 A 0.23 RIF
2 B 0.20 TMP
3 B 0.11 TMP
```
If the conditions to annotate are in a plate-shaped format, you can melt them into a columnar format before joining.
```python
>>> import pandas as pd
>>> import numpy as np
>>> a = pd.DataFrame(index=list("ABCDEFGH"),
... columns=range(1, 13),
... data=np.arange(1, 97).reshape(8, 12))
>>> a
1 2 3 4 5 6 7 8 9 10 11 12
A 1 2 3 4 5 6 7 8 9 10 11 12
B 13 14 15 16 17 18 19 20 21 22 23 24
C 25 26 27 28 29 30 31 32 33 34 35 36
D 37 38 39 40 41 42 43 44 45 46 47 48
E 49 50 51 52 53 54 55 56 57 58 59 60
F 61 62 63 64 65 66 67 68 69 70 71 72
G 73 74 75 76 77 78 79 80 81 82 83 84
H 85 86 87 88 89 90 91 92 93 94 95 96
>>> hts.pivot_plate(a, value_name="well_number")
row_id column_id well_number well_id plate_id
0 A 1 1 A01
1 B 1 13 B01
2 C 1 25 C01
3 D 1 37 D01
4 E 1 49 E01
.. ... ... ... ... ...
91 D 12 48 D12
92 E 12 60 E12
93 F 12 72 F12
94 G 12 84 G12
95 H 12 96 H12
[96 rows x 5 columns]
```
This also works on the multi-sheet dictionary output of `pd.read_excel(..., sheet_names=None)`.
```python
>>> hts.pivot_plate({'sheet_1': a}, value_name="well_number")
row_id column_id well_number well_id plate_id
0 A 1 1 A01 sheet_1
1 B 1 13 B01 sheet_1
2 C 1 25 C01 sheet_1
3 D 1 37 D01 sheet_1
4 E 1 49 E01 sheet_1
.. ... ... ... ... ...
91 D 12 48 D12 sheet_1
92 E 12 60 E12 sheet_1
93 F 12 72 F12 sheet_1
94 G 12 84 G12 sheet_1
95 H 12 96 H12 sheet_1
[96 rows x 5 columns]
```
Replicates within condition groups can be annotated.
```python
>>> import pandas as pd
>>> a = pd.DataFrame(dict(group=['g1', 'g1', 'g2', 'g2'],
... control=['n', 'n', 'p', 'p'],
... m_abs_ch1=[.1, .2, .9, .8],
... abs_ch1_wavelength=['600nm'] * 4))
>>> a
group control m_abs_ch1 abs_ch1_wavelength
0 g1 n 0.1 600nm
1 g1 n 0.2 600nm
2 g2 p 0.9 600nm
3 g2 p 0.8 600nm
>>> hts.replicate_table(a, group='group')
group control m_abs_ch1 abs_ch1_wavelength replicate
0 g1 n 0.1 600nm 1
1 g1 n 0.2 600nm 2
2 g2 p 0.9 600nm 2
3 g2 p 0.8 600nm 1
```
If you prefer, you can get a "wide" output.
```python
>>> hts.replicate_table(a, group='group', wide='m_abs_ch1')
replicate rep_1 rep_2
group
g1 0.2 0.1
g2 0.8 0.9
```
Values can be normalized to values between 0 and 1 relative to their positive (0%) and negative (100%) controls, optinally within groups or batches.
```python
>>> import pandas as pd
>>> a = pd.DataFrame(dict(control=['n', 'n', '', '', 'p', 'p'],
... m_abs_ch1=[.1, .2, .5, .4, .9, .8],
... abs_ch1_wavelength=['600nm'] * 6))
>>> a
control m_abs_ch1 abs_ch1_wavelength
0 n 0.1 600nm
1 n 0.2 600nm
2 0.5 600nm
3 0.4 600nm
4 p 0.9 600nm
5 p 0.8 600nm
>>> hts.normalize(a, control_col='control', pos='p', neg='n', measurement_col='m_abs_ch1')
control m_abs_ch1 abs_ch1_wavelength m_abs_ch1_neg_mean m_abs_ch1_pos_mean m_abs_ch1_norm
0 n 0.1 600nm 0.15 0.85 1.071429
1 n 0.2 600nm 0.15 0.85 0.928571
2 0.5 600nm 0.15 0.85 0.500000
3 0.4 600nm 0.15 0.85 0.642857
4 p 0.9 600nm 0.15 0.85 -0.071429
5 p 0.8 600nm 0.15 0.85 0.071429
```
The scaling can be reversed with `flip=True`.
```python
>>> hts.normalize(a, control_col='control', pos='p', neg='n', measurement_col='m_abs_ch1', flip=True)
control m_abs_ch1 abs_ch1_wavelength m_abs_ch1_neg_mean m_abs_ch1_pos_mean m_abs_ch1_norm
0 n 0.1 600nm 0.15 0.85 -0.071429
1 n 0.2 600nm 0.15 0.85 0.071429
2 0.5 600nm 0.15 0.85 0.500000
3 0.4 600nm 0.15 0.85 0.357143
4 p 0.9 600nm 0.15 0.85 1.071429
5 p 0.8 600nm 0.15 0.85 0.928571
```
Summary statstics and statsitcial tests relative to the negative controls can be generated.
```python
>>> a = pd.DataFrame(dict(gene=['g1', 'g1', 'g2', 'g2', 'g1', 'g1', 'g2', 'g2'],
... compound=['n', 'n', 'n', 'n', 'cmpd1', 'cmpd1', 'cmpd2', 'cmpd2'],
... m_abs_ch1=[.1, .2, .9, .8, .1, .3, .5, .45],
... abs_ch1_wavelength=['600nm'] * 8))
>>> a
gene compound m_abs_ch1 abs_ch1_wavelength
0 g1 n 0.10 600nm
1 g1 n 0.20 600nm
2 g2 n 0.90 600nm
3 g2 n 0.80 600nm
4 g1 cmpd1 0.10 600nm
5 g1 cmpd1 0.30 600nm
6 g2 cmpd2 0.50 600nm
7 g2 cmpd2 0.45 600nm
>>> hts.summarize(a, measurement_col='m_abs_ch1', control_col='compound', neg='n', group='gene')
gene abs_ch1_wavelength m_abs_ch1_mean m_abs_ch1_std ... m_abs_ch1_t.stat m_abs_ch1_t.p m_abs_ch1_ssmd m_abs_ch1_log10fc
0 g1 600nm 0.1750 0.095743 ... 0.361158 0.742922 0.210042 0.066947
1 g2 600nm 0.6625 0.221265 ... -1.544396 0.199787 -0.807183 -0.108233
[2 rows x 12 columns]
>>> hts.summarize(a, measurement_col='m_abs_ch1', control_col='compound', neg='n', group=['gene', 'compound'])
gene compound abs_ch1_wavelength m_abs_ch1_mean ... m_abs_ch1_t.stat m_abs_ch1_t.p m_abs_ch1_ssmd m_abs_ch1_log10fc
0 g1 n 600nm 0.150 ... 0.000000 1.000000 0.000000 0.000000
1 g2 n 600nm 0.850 ... 0.000000 1.000000 0.000000 0.000000
2 g1 cmpd1 600nm 0.200 ... 0.447214 0.711723 0.316228 0.124939
3 g2 cmpd2 600nm 0.475 ... -6.708204 0.044534 -4.743416 -0.252725
[4 rows x 13 columns]
```
## Documentation
Full API documentation is at [ReadTheDocs](https://hts-tools.readthedocs.org).
Raw data
{
"_id": null,
"home_page": "",
"name": "hts-tools",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "",
"keywords": "science,assay,platereader,analysis",
"author": "",
"author_email": "Eachan Johnson <eachan.johnson@crick.ac.uk>",
"download_url": "https://files.pythonhosted.org/packages/e6/cf/0ffdf62a5604dc816862c322ccaa4e23b253f008df78f44c904a7959b3c5/hts-tools-0.0.2.post0.tar.gz",
"platform": null,
"description": "# \ud83c\udfee hts-tools\n\n\n\n\n\nParsing and analysing platereader absorbance and fluorescence data.\n\n- [Installation](#installation)\n- [Command-line usage](#command-line-usage)\n - [First pipeline example](#first-pipeline-example)\n - [Parsing platereader exports](#parsing-platereader-exports)\n - [Adding experimental conditions](#adding-experimental-conditions)\n - [Normalization within batches](#normalization-within-batches)\n - [Plotting dose responses](#plotting-dose-response)\n - [Second pipeline example](#second-pipeline-example)\n - [Converting plate-shaped data to columns](#converting-plate-shaped-data-to-columns)\n - [Statistcal testing](#statistical-testing)\n - [Other commands](#other-commands)\n- [Python API](#python-api)\n- [Documentation](#documentation)\n\n## Installation\n\n### The easy way\n\nInstall the pre-compiled version from PyPI:\n\n```bash\npip install hts-tools\n```\n\n### From source\n\nClone the repository, then `cd` into it. Then run:\n\n```bash\npip install -e .\n```\n\n## Command-line usage\n\n**hts-tools** provides command-line utlities to analyse and plot data\nfrom platereaders, starting with the _raw_ exported data with no \nmanual conversion or copy-pasting needed. The tools complete specific tasks which \ncan be easily composed into analysis pipelines, because the TSV table output goes to\n`stdout` by default so they can be piped from one tool to another.\n\nTo get a list of commands (tools), do\n\n```bash\nhts --help\n```\n\nAnd to get help for a specific command, do\n\n```bash\nhts <command> --help\n```\n\nFor the Python API, [see below](#python-api).\n\n## First pipeline example\n\nThis command takes several exported Excel files (matching the pattern `plate-?.xlsx`) from a Biotek platereader, \nadds annotations on experimental conditions, and normalizes the data based on positive and negative controls. \nFinally, dose-response curves are plotted. \n\n```bash\nhts parse plate-?.xlsx --data-shape row --vendor Biotek \\\n | hts join --right layout.xlsx \\\n | hts normalize --control compound_name \\\n --positive RIF --negative DMSO \\\n --grouping plate_id --method npg \\\n | hts plot-dose -x concentration --facets guide_name --color compound_name \\\n --output plt-test\n```\n\n### Parsing platereader exports\n\nThe command `hts parse` converts Excel or CSV or TSV files exported from platereader software from a specified vendor into a \nuniform columnar table format used by all downstream hts-tools. The `data-shape` option indicates whether the export was in plate \nformat or row-wise table format. It should usually be the first command in a pipeline. \n\n`hts parse` produces a table with at least the following columns (with example entries):\n\n| row_id\t| column_id\t| well_id\t | plate_id\t| data_filename\t| data_sheet | measured_abs_ch1 | abs_ch1_wavelength |\n| ------- | --------- | -------- | -------- | ------------- | ---------------- | ---------------- | ------------------ |\n| A | 1 | A01 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.366 | 600nm |\n| A | 2 | A02 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.402\t | 600nm |\n\nIf only fluorescence was measured, the last two columns would be called `measured_fluor_ch1` and `fluor_ch1_wavelength`. If there were multiple\nmeasurements, then they will appear as additional columns with increasing `ch` (channel) numbers, for example, `measured_fluor_ch2`, `fluor_ch2_wavelength`.\n\nThere will also be additional columns using information scraped from the input file. These will usually have headings starting with `meta_`,\nsuch as `meta_protocol_file_path`, `meta_date`, `meta_time`, and `meta_reader_serial_number`.\n\n### Adding experimental conditions\n\nThe output from `hts` parse is passed to `hts join`, which combines two tables based on values in shared columns (known as a \ndatabase join). For example, if `layout.xlsx` contains a sheet with this table:\n\n| column_id\t| plate_id | compound_name |\n| --------- | -------- | ------------- |\n| 2 | Plate 6 | trimethoprim |\n| 1 | Plate 6 | moxifloxacin |\n\nthen `hts parse plate-?.xlsx --data-shape row | hts join --right layout.xlsx` will result in \n\n| row_id\t| column_id\t| well_id\t | plate_id\t| data_filename\t| data_sheet | measured_abs_ch1 | abs_ch1_wavelength | compound_name |\n| ------- | --------- | -------- | -------- | ------------- | ---------------- | ---------------- | ------------------ | ------------- |\n| A | 1 | A01 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.366 | 600nm | moxifloxacin |\n| A | 2 | A02 | Plate 6 | plate-6.xlsx | Plate6 - Sheet 1 | 0.402\t | 600nm | trimethoprim |\n\nSince `column_id` and `plate_id` are the only column headings in common, the entries of these columns are used to match the rows of the \ntables. So where `column_id = 2`, `compound_name = trimethoprim` will be added. \n\nIf you join an Excel XLSX file containing multiple sheets, these will each be joined in order. In this way, you can add experimental\nconditions easily by, for example, first joining the conditions that vary by plate and row (such as compound), then by column \n(such as concentration). This approach is very flexible, and you can join on any number of columns and add any new ones you like as\nlong as the column headings aren't repeated. \n\n### Normalization within batches \n\nThe `hts normalize` command normalizes raw measured data based on controls. This can add power to \ndownstream statsical analyses by mitigating batch-to-batch variation. \n\n`hts normalize` adds new columns for each measured column. These columns start with `calc_` and end with `_norm.{method}`, for example `calc_abs_ch1_norm.npg` and `calc_abs_ch2_norm.npg`. Two methods are offered which \noptionally normalize within within groups, such as batches or plates.\n\n#### Normalized proportion of growth (NPG)\n \nThis method scales raw data to be between $0$ and $1$ based on positive _and_ negative \ncontrols, optionally within groups (or batches) of measurements. In the example above, \nthe positive and negative controls are defined as `RIF` and `DMSO`, and should be found in the column `compound_name` (which may have been added by `hts join`).\n\nThe positive and negative controls are averaged within each value in the `--grouping` column. In \nthe example above, they will be averaged for each `plate_id`, and these will be used to normalize \nthe measured values of that `plate_id` according to:\n\n$$s = \\frac{m - \\mu_p}{\\mu_n - \\mu_p}$$\n\nwhere $s$ is the normalized value, $m$ is the measured value, and $\\mu_p$ and $\\mu_n$ are the mean positive and negative controls. \n\n#### Proportion of negative (PON) [default]\n\nThis method scales raw data relative negative controls _only_, optionally within groups (or batches) of measurements. \n\nThe negative controls are averaged within each value in the `--grouping` column. In \nthe example above, they would be averaged for each `plate_id`, and these will be used to normalize \nthe measured values of that `plate_id` according to:\n\n$$s = \\frac{m}{\\mu_n}$$\n\nwhere $s$ is the normalized value, $m$ is the measured value, and $\\mu_n$ is the mean negative control. \n\n### Plotting dose response\n\n`hts plot-dose` is a very flexible command which takes the columnar tables as input and plots the data in almost any breakdown using a color-blind \npalette. The required `-x` option indicates which column to use as the x-axis (usually concentration). The y-axis will be values in all the measured \nand calculated columns (`hts plot-dr` plots them all automatically in seaparte files). The other options allow splitting the plots by file, facet \n(panel) and color according to the values in columns. \n\nThe example above, `hts plot-dose -x concentration --facets guide_name --color compound_name`, will produce plots like this:\n\n<img src=\"docs/source/_static/plt-test_calc_abs_ch1_norm_.png\" alt=\"\" width=\"600\">\n\nThe panels each value with the same `guide_name` is in a `facet` (panel), and the lines are colored by `compound_name`.\n\n## Second pipeline example\n\nHere is another example showing the sequential use of `hts join` to join two tables of experimental data, and two other commands: `hts pivot` and \n`hts summarize`.\n\n```bash\nhts pivot compounds.xlsx \\\n --name compound_name\n\nhts parse plate-*.txt --data-shape plate \\\n | hts join --right sample-sheet.csv \\\n | hts join --right pivoted-compounds.tsv \\\n | hts normalize --control compound_name --positive RIF --negative DMSO --grouping strain_name plate_id \\\n | hts summarize --control compound_name --positive RIF --negative DMSO --grouping strain_name compound_name --plot summary \\\n > summary.tsv\n```\n\nThese commands are explained below.\n\n### Converting plate shaped data to columns\n\nSometimes you will want to use data, such as plate layouts, which are in a plate-shaped layout instead of a column format. For example:\n\n<img src=\"docs/source/_static/plate-screenshot.png\" alt=\"\" width=\"800\">\n\nYou can convert this to column format using `hts pivot`, which produces a table in the following format:\n\n| row_id | column_id | compound_name | well_id | plate_id | filename |\n| ------ | --------- | ------------- | ------- | ------------------------ | ------------------------ |\n| C\t | 2\t | RIF\t | C02\t | Plate 7\t | compounds.xlsx |\n| D\t | 2\t | LYSINE\t | D02\t | Plate 7\t | compounds.xlsx |\n| E\t | 2\t | DMSO\t | E02\t | Plate 7\t | compounds.xlsx |\n| F\t | 2\t | RIF\t | F02\t | Plate 7\t | compounds.xlsx |\n\nIt is assumed that there is one plate per sheet for Excel files, and one plate per file for TSV and CSV files. The\nplate name is taken from the sheet name (Excel) or filename (other formats).\n\nYou can prepend the names of the `plate_id` and `filename` columns with the `-x` option. for example, \n`hts pivot compounds.xlsx -x compound_source --name compound_name` would have columns `compound_source_plate_id`\n`compound_source_filename`. This is helpful when usign `hts join` later where the plate and filename \ncolumns would otherwise be shared but have different meanings and values, and you don't want to accidentally \njoin on them.\n\n### Statistical testing\n\nGroups of values (such as replicates) can be compared against a negative control for statistical testing using\n`hts summarize`. The `--grouping` option indicates the columns whose values together indicate values which are\nreplicates of a particular condition of interest. For example, `--grouping strain_name compound_name` would\nindicate that values which have the same `strain_name` and `compound_name` are replicates.\n\nStatistical tests compare to the `--negative` values, and use all measured and normalized (`calc_*_norm`) columns.\nCurrently, the Student's t-test and Mann-whitney U-test are implemented. The t-test is best suited to Normal-distributed\ndata while the MWU is better for other distributions which might not have a nice bell curve distribution.\n\nAlthough `hts summarize` calcualtes both tests simultaneously, it's not a good idea to look for \"significant\" $p$-values \nin both. This is called p-hacking, and leads to false positives. Instead, decide which test is most appropriate for your\ndata and stick with that one.\n\nThis command also calculates other summary statistics such as between-replicate mean, variance, and SSMD. If a filename\nprefix is provided to the `--plot` option, then volcano and flashlight plots are produced, which may be useful for identifying \nhits of high throughput screens.\n\n<img src=\"docs/source/_static/summary-counter_fluor_ch1_norm_volcano.png\" alt=\"\" width=\"600\">\n<img src=\"docs/source/_static/summary-counter_fluor_ch1_norm_flashlight.png\" alt=\"\" width=\"600\">\n\n## Other commands\n\nThere are several other commands from **hts-tools** which take as input the output from `hts parse`, joined\nto an experimental data table (`layout.xlsx` in these examples).\n\n- `hts qc`\n\nDo quality cpntrol checks by calculating mean, standard deviation, Z\\'-factor, and SSMD, and plotting them.\n\n```bash\nhts parse plate-?.xlsx --data-shape row \\\n | hts join --right layout.xlsx \\\n | hts qc --control compound_name --positive RIF --negative DMSO --grouping strain_name plate_id --plot qc-plot \\\n > qc.tsv\n```\n\n<img src=\"docs/source/_static/qc-plot_fluor_ch1_mean_sd.png\" alt=\"\" width=\"600\">\n<img src=\"docs/source/_static/qc-plot_fluor_ch1_zprime.png\" alt=\"\" width=\"600\">\n\n- `hts plot-hm`\n\nPlot heatmaps of signal intensity arranged by plate well. This can be useful to identify unwanted within-plate variability.\n\n```bash\nhts parse plate-?.xlsx --data-shape row \\\n | hts join --right layout.xlsx \\\n | hts plot-hm --grouping strain_name sample_id plate_id --output hm\n```\n\nHere, `--grouping` identifies the columns which indicate values coming from the same plate. One file is produced per\nmeasured and normalized (`calc_*_norm`) column.\n\n<img src=\"docs/source/_static/hm_calc_fluor_ch1_norm_heatmap.png\" alt=\"\" width=\"600\">\n\n- `hts plot-rep`\n\nPlot two replicates against each other for each condition.\n\n```bash\nhts parse plate-?.xlsx --data-shape row \\\n | hts join --right layout.xlsx \\\n | hts plot-rep --control compound_name --positive RIF --negative DMSO --grouping strain_name compound_name --output rep\n```\n\nHere, `--grouping` identifies the unique conditions within which values are treated as replicates. The positives and negatives\nare plotted as different colors. One file is produced per measured and normalized (`calc_*_norm`) column.\n\n<img src=\"docs/source/_static/rep_measured_fluor_ch1_replicates.png\" alt=\"\" width=\"600\">\n\nIn the plots, the left column is on a linear-linear scale and the right column is on a log-log scale. There is one row\nof plots per wavelength set in the dataset.\n\n- `hts plot-hist`\n\nPlot histograms of the data values.\n\n```bash\nhts parse plate-?.xlsx --data-shape row \\\n | hts join --right layout.xlsx \\\n | hts plot-hist --control compound_name --positive RIF --negative DMSO --output hist\n```\n\nThe positives and negatives are plotted as different colors. One file is produced per measured and normalized (`calc_*_norm`) column.\n\n<img src=\"docs/source/_static/hist_measured_fluor_ch1_histogram.png\" alt=\"\" width=\"600\">\n\nIn the plots, the left two columns are on a linear-linear scale and the right two columns are on a log-log scale. There is one row\nof plots per wavelength set in the dataset.\n\n## Python API\n\n**hts-tools** can be imported into Python to help make custom analyses.\n\n```python\n>>> import htstools as hts\n```\n\nYou can read raw exports from platereader software into a columnar Pandas dataframe.\n\n```python\n>>> hts.from_platereader(\"plates.xlsx\", shape=\"plate\", vendor=\"Biotek\")\n```\n\nOnce in the columnar format, you can annotate experimental conditions.\n\n```python\n>>> import pandas as pd\n>>> a = pd.DataFrame(dict(column=['A', 'B', 'A', 'B'], \n... abs=[.1, .2, .23, .11]))\n>>> a \n column abs\n0 A 0.10\n1 B 0.20\n2 A 0.23\n3 B 0.11\n>>> b = pd.DataFrame(dict(column=['B', 'A'], \n... drug=['TMP', 'RIF']))\n>>> b \n column drug\n0 B TMP\n1 A RIF\n>>> shared_cols, data = join(a, b)\n>>> shared_cols\n('column',)\n>>> data \ncolumn abs drug\n0 A 0.10 RIF\n1 A 0.23 RIF\n2 B 0.20 TMP\n3 B 0.11 TMP\n```\n\nIf the conditions to annotate are in a plate-shaped format, you can melt them into a columnar format before joining.\n\n```python\n>>> import pandas as pd\n>>> import numpy as np\n>>> a = pd.DataFrame(index=list(\"ABCDEFGH\"), \n... columns=range(1, 13), \n... data=np.arange(1, 97).reshape(8, 12))\n>>> a \n 1 2 3 4 5 6 7 8 9 10 11 12\nA 1 2 3 4 5 6 7 8 9 10 11 12\nB 13 14 15 16 17 18 19 20 21 22 23 24\nC 25 26 27 28 29 30 31 32 33 34 35 36\nD 37 38 39 40 41 42 43 44 45 46 47 48\nE 49 50 51 52 53 54 55 56 57 58 59 60\nF 61 62 63 64 65 66 67 68 69 70 71 72\nG 73 74 75 76 77 78 79 80 81 82 83 84\nH 85 86 87 88 89 90 91 92 93 94 95 96\n>>> hts.pivot_plate(a, value_name=\"well_number\") \n row_id column_id well_number well_id plate_id\n0 A 1 1 A01 \n1 B 1 13 B01 \n2 C 1 25 C01 \n3 D 1 37 D01 \n4 E 1 49 E01 \n.. ... ... ... ... ...\n91 D 12 48 D12 \n92 E 12 60 E12 \n93 F 12 72 F12 \n94 G 12 84 G12 \n95 H 12 96 H12 \n\n[96 rows x 5 columns]\n```\n\nThis also works on the multi-sheet dictionary output of `pd.read_excel(..., sheet_names=None)`.\n\n```python\n>>> hts.pivot_plate({'sheet_1': a}, value_name=\"well_number\") \nrow_id column_id well_number well_id plate_id\n0 A 1 1 A01 sheet_1\n1 B 1 13 B01 sheet_1\n2 C 1 25 C01 sheet_1\n3 D 1 37 D01 sheet_1\n4 E 1 49 E01 sheet_1\n.. ... ... ... ... ...\n91 D 12 48 D12 sheet_1\n92 E 12 60 E12 sheet_1\n93 F 12 72 F12 sheet_1\n94 G 12 84 G12 sheet_1\n95 H 12 96 H12 sheet_1\n\n[96 rows x 5 columns]\n```\n\nReplicates within condition groups can be annotated.\n\n```python\n>>> import pandas as pd\n>>> a = pd.DataFrame(dict(group=['g1', 'g1', 'g2', 'g2'], \n... control=['n', 'n', 'p', 'p'], \n... m_abs_ch1=[.1, .2, .9, .8], \n... abs_ch1_wavelength=['600nm'] * 4))\n>>> a \n group control m_abs_ch1 abs_ch1_wavelength\n0 g1 n 0.1 600nm\n1 g1 n 0.2 600nm\n2 g2 p 0.9 600nm\n3 g2 p 0.8 600nm\n>>> hts.replicate_table(a, group='group') \n group control m_abs_ch1 abs_ch1_wavelength replicate\n0 g1 n 0.1 600nm 1\n1 g1 n 0.2 600nm 2\n2 g2 p 0.9 600nm 2\n3 g2 p 0.8 600nm 1\n```\n\nIf you prefer, you can get a \"wide\" output.\n\n```python\n>>> hts.replicate_table(a, group='group', wide='m_abs_ch1') \nreplicate rep_1 rep_2\ngroup \ng1 0.2 0.1\ng2 0.8 0.9\n```\n\nValues can be normalized to values between 0 and 1 relative to their positive (0%) and negative (100%) controls, optinally within groups or batches.\n\n```python\n>>> import pandas as pd\n>>> a = pd.DataFrame(dict(control=['n', 'n', '', '', 'p', 'p'], \n... m_abs_ch1=[.1, .2, .5, .4, .9, .8], \n... abs_ch1_wavelength=['600nm'] * 6))\n>>> a \n control m_abs_ch1 abs_ch1_wavelength\n0 n 0.1 600nm\n1 n 0.2 600nm\n2 0.5 600nm\n3 0.4 600nm\n4 p 0.9 600nm\n5 p 0.8 600nm\n>>> hts.normalize(a, control_col='control', pos='p', neg='n', measurement_col='m_abs_ch1') \n control m_abs_ch1 abs_ch1_wavelength m_abs_ch1_neg_mean m_abs_ch1_pos_mean m_abs_ch1_norm\n0 n 0.1 600nm 0.15 0.85 1.071429\n1 n 0.2 600nm 0.15 0.85 0.928571\n2 0.5 600nm 0.15 0.85 0.500000\n3 0.4 600nm 0.15 0.85 0.642857\n4 p 0.9 600nm 0.15 0.85 -0.071429\n5 p 0.8 600nm 0.15 0.85 0.071429\n```\n\nThe scaling can be reversed with `flip=True`.\n\n```python\n>>> hts.normalize(a, control_col='control', pos='p', neg='n', measurement_col='m_abs_ch1', flip=True) \n control m_abs_ch1 abs_ch1_wavelength m_abs_ch1_neg_mean m_abs_ch1_pos_mean m_abs_ch1_norm\n0 n 0.1 600nm 0.15 0.85 -0.071429\n1 n 0.2 600nm 0.15 0.85 0.071429\n2 0.5 600nm 0.15 0.85 0.500000\n3 0.4 600nm 0.15 0.85 0.357143\n4 p 0.9 600nm 0.15 0.85 1.071429\n5 p 0.8 600nm 0.15 0.85 0.928571\n```\n\nSummary statstics and statsitcial tests relative to the negative controls can be generated.\n\n```python\n\n>>> a = pd.DataFrame(dict(gene=['g1', 'g1', 'g2', 'g2', 'g1', 'g1', 'g2', 'g2'], \n ... compound=['n', 'n', 'n', 'n', 'cmpd1', 'cmpd1', 'cmpd2', 'cmpd2'], \n ... m_abs_ch1=[.1, .2, .9, .8, .1, .3, .5, .45], \n ... abs_ch1_wavelength=['600nm'] * 8))\n >>> a \n gene compound m_abs_ch1 abs_ch1_wavelength\n 0 g1 n 0.10 600nm\n 1 g1 n 0.20 600nm\n 2 g2 n 0.90 600nm\n 3 g2 n 0.80 600nm\n 4 g1 cmpd1 0.10 600nm\n 5 g1 cmpd1 0.30 600nm\n 6 g2 cmpd2 0.50 600nm\n 7 g2 cmpd2 0.45 600nm\n >>> hts.summarize(a, measurement_col='m_abs_ch1', control_col='compound', neg='n', group='gene') \n gene abs_ch1_wavelength m_abs_ch1_mean m_abs_ch1_std ... m_abs_ch1_t.stat m_abs_ch1_t.p m_abs_ch1_ssmd m_abs_ch1_log10fc\n 0 g1 600nm 0.1750 0.095743 ... 0.361158 0.742922 0.210042 0.066947\n 1 g2 600nm 0.6625 0.221265 ... -1.544396 0.199787 -0.807183 -0.108233\n\n [2 rows x 12 columns]\n >>> hts.summarize(a, measurement_col='m_abs_ch1', control_col='compound', neg='n', group=['gene', 'compound'])\n gene compound abs_ch1_wavelength m_abs_ch1_mean ... m_abs_ch1_t.stat m_abs_ch1_t.p m_abs_ch1_ssmd m_abs_ch1_log10fc\n 0 g1 n 600nm 0.150 ... 0.000000 1.000000 0.000000 0.000000\n 1 g2 n 600nm 0.850 ... 0.000000 1.000000 0.000000 0.000000\n 2 g1 cmpd1 600nm 0.200 ... 0.447214 0.711723 0.316228 0.124939\n 3 g2 cmpd2 600nm 0.475 ... -6.708204 0.044534 -4.743416 -0.252725\n\n [4 rows x 13 columns]\n```\n\n## Documentation\n\nFull API documentation is at [ReadTheDocs](https://hts-tools.readthedocs.org).\n",
"bugtrack_url": null,
"license": "MIT License Copyright (c) [year] [fullname] Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.",
"summary": "Parsing and analysing platereader absorbance and fluorescence data.",
"version": "0.0.2.post0",
"project_urls": {
"Bug Tracker": "https://github.com/scbirlab/hts-tools/issues",
"Homepage": "https://github.com/scbirlab/hts-tools"
},
"split_keywords": [
"science",
"assay",
"platereader",
"analysis"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "7045230f65843450c06d04179d7563cee7e457a9d50e26d067fb68c55561c522",
"md5": "1ed3101b361368733f81da634048e768",
"sha256": "1a4b0a2bb23a8b682f65172237ddb6df64901cf0a770a4cd0f30256df50f3333"
},
"downloads": -1,
"filename": "hts_tools-0.0.2.post0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "1ed3101b361368733f81da634048e768",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 35432,
"upload_time": "2023-11-16T14:25:22",
"upload_time_iso_8601": "2023-11-16T14:25:22.237528Z",
"url": "https://files.pythonhosted.org/packages/70/45/230f65843450c06d04179d7563cee7e457a9d50e26d067fb68c55561c522/hts_tools-0.0.2.post0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e6cf0ffdf62a5604dc816862c322ccaa4e23b253f008df78f44c904a7959b3c5",
"md5": "5a366fe9b665eaffd9e60a8a73cc88ff",
"sha256": "31b7f11b2e065aa113bdaf1fc32ec5669914cf1e63cfe4c5499bb0558853bdc5"
},
"downloads": -1,
"filename": "hts-tools-0.0.2.post0.tar.gz",
"has_sig": false,
"md5_digest": "5a366fe9b665eaffd9e60a8a73cc88ff",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 32583,
"upload_time": "2023-11-16T14:25:29",
"upload_time_iso_8601": "2023-11-16T14:25:29.118697Z",
"url": "https://files.pythonhosted.org/packages/e6/cf/0ffdf62a5604dc816862c322ccaa4e23b253f008df78f44c904a7959b3c5/hts-tools-0.0.2.post0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-11-16 14:25:29",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "scbirlab",
"github_project": "hts-tools",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "hts-tools"
}
