hitseekpy

reference

List et al. (2016). Comprehensive analysis of high-throughput screens with HiTSeekR, Nucleic Acids Research, Volume 44, Issue 14 (https://doi.org/10.1093/nar/gkw554)

Install

pip install hitseekpy

How to use

see also: https://davegrays.github.io/hitseekpy/

import pandas as pd

from hitseekpy.proc import Plate
from hitseekpy.scores import get_b_scores, get_z_scores

prepare the data

# load plate data and make sure wells are properly labeled
df = pd.read_csv("assay.1156.0009.tsv", delimiter="\t")
df[["WELL", "RAW_VALUE_A"]]

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	WELL	RAW_VALUE_A
0	A01	1960.0
1	A02	1400.0
2	A03	1600.0
3	A04	1560.0
4	A05	1640.0
...	...	...
379	P20	1800.0
380	P21	1240.0
381	P22	1520.0
382	P23	1200.0
383	P24	1480.0

384 rows × 2 columns

# extract well values into a 2d array
p = Plate(df, n_rows=16, n_cols=24, well_column="WELL", value_column="RAW_VALUE_A")

# check that they match the df
print(p.vals[0, :5])
print(p.vals[-1, -5:])

[1960. 1400. 1600. 1560. 1640.]
[1800. 1240. 1520. 1200. 1480.]

show z-scores

z_scores = get_z_scores(p.vals)
# show just the first 5 rows and columns
z_scores[0:5, 0:5]

array([[ 2.38139535, -0.39689922,  0.59534884,  0.39689922,  0.79379845],
       [-0.79379845,  0.99224806, -0.79379845, -1.19069767, -0.19844961],
       [ 1.19069767, -0.99224806,  0.        , -0.19844961,  0.99224806],
       [-0.99224806,  0.        ,  0.99224806,  0.19844961,  0.99224806],
       [-0.39689922,  1.98449612,  0.        , -0.39689922,  2.77829457]])

show b-scores

b_scores_dict = get_b_scores(p.vals)

b_scores_dict["row_effect"]

array([  38.359375 ,  -24.0625   , -101.40625  ,  -92.1484375,
         51.171875 ,    5.3125   ,   -2.6171875,  -89.6484375,
         42.1875   ,   38.359375 ,    2.1875   ,   73.9453125,
          8.125    ,  146.5234375,  -73.359375 ,  -17.6953125,
         47.9296875,   76.6015625,  -66.5234375,  -46.1328125,
        -65.       , -184.9609375, -132.1875   ,   27.8125   ])

b_scores_dict["column_effect"]

array([  50.6640625, -177.2265625,   55.8984375,  102.8515625,
         88.2421875,  -23.8671875,    8.4765625,   -2.8515625,
         57.3046875,    8.4765625,    2.8515625,   -7.8515625,
        -69.1796875, -106.6796875,  -97.1484375,  -77.7734375])

# show just the first 5 rows and columns
b_scores_dict["normalized"][0:5, 0:5]

array([[ 2.26025391e+00, -8.49609375e-01,  8.83789062e-01,
         5.75927734e-01,  1.80175781e-01],
       [-3.15429688e-01,  2.32470703e+00,  5.58105469e-01,
         2.44140625e-04,  3.54492188e-01],
       [ 7.27539062e-01, -1.63232422e+00,  1.01074219e-01,
        -2.06787109e-01,  3.97460938e-01],
       [-2.31591797e+00, -6.75781250e-01,  1.05761719e+00,
        -2.44140625e-04,  1.04003906e-01],
       [-1.47460938e+00,  1.91552734e+00, -1.01074219e-01,
        -6.58935547e-01,  2.44531250e+00]])