{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# HEWL S-SAD Merging Statistics\n", "\n", "Merging statistics are a useful means to assess data quality in crystallography. However, each statistic has inherent shortcomings. For example, R-merge will appear inflated if the multiplicity is high, and the Pearson correlation coefficients used for $CC_{1/2}$ are very sensitive to outliers. \n", "\n", "Most scaling and merging programs output multiple merging statistics to get around these shortcomings. However, one can imagine that it could also be useful to customize certain parameters, such as how many resolution bins are used. Or, perhaps a better statistic will be developed that is worth implementing. \n", "\n", "In this notebook, ``reciprocalspaceship`` is used to compute half-dataset correlation coefficients ($CC_{1/2}$ and $CC_{anom}$) for a dataset collected from a tetragonal hen egg-white lysozyme (HEWL) crystal at 6550 eV. \n", "These data are unmerged, but were scaled in AIMLESS. \n", "They contain sufficient sulfur anomalous signal to determine a solution by the SAD method. \n", "This illustrates the use of ``rs`` to implement a merging routine, create a custom analysis, and could be useful as a template for other exploratory crystallographic data analyses. \n", "As an example, we will compare half-dataset correlations computed using both Pearson and Spearman correlation coefficients." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "sns.set_context(\"notebook\", font_scale=1.3)\n", "import numpy as np" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import reciprocalspaceship as rs" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "0.9.9\n" ] } ], "source": [ "print(rs.__version__)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Load scaled, unmerged data \n", "\n", "This data has been scaled in AIMLESS. The data includes the image number and the scaled **I** and **SIGI** values. " ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "hewl = rs.read_mtz(\"data/HEWL_unmerged.mtz\")" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
BATCHISIGIPARTIAL
HKL
004137696.521287.83294False
4520710.681288.107025False
4856672.0563487.75671False
41239642.4748587.90302False
42160655.7178387.74394False
\n", "
" ], "text/plain": [ " BATCH I SIGI PARTIAL\n", "H K L \n", "0 0 4 137 696.5212 87.83294 False\n", " 4 520 710.6812 88.107025 False\n", " 4 856 672.05634 87.75671 False\n", " 4 1239 642.47485 87.90302 False\n", " 4 2160 655.71783 87.74394 False" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "hewl.head()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Number of observed reflections: 816804\n" ] } ], "source": [ "print(f\"Number of observed reflections: {len(hewl)}\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Merging with Inverse-Variance Weights\n", "\n", "Since the input data are unmerged, we will implement the inverse-variance weighting scheme used by AIMLESS to merge the observations. \n", "The weighted average is a better estimator of the true mean than the raw average, and this weighting scheme corresponds to the maximum likelihood estimator of the true mean if we assume that the observations are normally-distributed about the true mean. \n", "\n", "The merged intensity for each reflection, $I_h$, can be determined from the observed intensities, $I_{h,i}$, and error estimates, $\\sigma_{h,i}$, as follows:\n", "\n", "\\begin{equation}\n", "I_h = \\frac{\\sum_{i}w_{h,i} I_{h,i}}{\\sum_{i} w_{h,i}}\n", "\\end{equation}\n", "\n", "where the weight for each observation, $w_{h,i}$ is given by:\n", "\n", "\\begin{equation}\n", "w_{h,i} = \\frac{1}{(\\sigma_{h,i})^2}\n", "\\end{equation}\n", "\n", "The updated estimate of the uncertainty, $\\sigma_{h}$, is given by:\n", "\n", "\\begin{equation}\n", "\\sigma_{h} = \\sqrt{\\frac{1}{\\sum_{i} w_{h,i}}}\n", "\\end{equation}\n", "\n", "Let's start by implementing the above equations in a function that will compute the merged $I_h$ and $\\sigma_h$. We will use the [Pandas groupby](https://pandas.pydata.org/pandas-docs/stable/user_guide/groupby.html) methods to apply this function across the unique Miller indices in the ``DataSet``. If we group Friedel pairs together, we will refer to the quantity as **IMEAN**, and if we keep the Friedel pairs separate we will refer to the quantities as **I(+)** and **I(-)**." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "def merge(dataset, anomalous=False):\n", " \"\"\"\n", " Merge dataset using inverse-variance weights.\n", " \n", " Parameters\n", " ----------\n", " dataset : rs.DataSet\n", " DataSet to be merged containing scaled I and SIGI\n", " anomalous : bool\n", " If True, I(+) and I(-) will be reported. If False,\n", " IMEAN will be reported\n", " \n", " Returns\n", " -------\n", " rs.DataSet\n", " Merged DataSet object\n", " \"\"\"\n", " ds = dataset.hkl_to_asu(anomalous=anomalous)\n", " ds[\"w\"] = ds['SIGI']**-2\n", " ds[\"wI\"] = ds[\"I\"] * ds[\"w\"]\n", " g = ds.groupby([\"H\", \"K\", \"L\"])\n", " \n", " result = g[[\"w\", \"wI\"]].sum()\n", " result[\"I\"] = result[\"wI\"] / result[\"w\"]\n", " result[\"SIGI\"] = np.sqrt(1 / result[\"w\"])\n", " result = result.loc[:, [\"I\", \"SIGI\"]]\n", " result.merged = True\n", " \n", " if anomalous:\n", " result = result.unstack_anomalous()\n", " \n", " return result" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Using `anomalous=False`, this function can be used to compute **IMEAN** and **SIGIMEAN** by including both Friedel pairs:" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "result1 = merge(hewl, anomalous=False)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
ISIGI
HKL
232291349.800722.878157
37190260.09969.075029
171115146.023242.895325
351149.6299983.902024
155187.8371151.039951
\n", "
" ], "text/plain": [ " I SIGI\n", "H K L \n", "23 22 9 1349.8007 22.878157\n", "37 19 0 260.0996 9.075029\n", "17 11 15 146.02324 2.895325\n", "35 1 14 9.629998 3.902024\n", "15 5 18 7.837115 1.039951" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result1.sample(5)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Using `anomalous=True`, this function can be used to compute **I(+)**, **SIGI(+)**, **I(-)**, and **SIGI(-)** by separating Friedel pairs:" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "result2 = merge(hewl, anomalous=True)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
I(+)SIGI(+)SIGI(-)I(-)
HKL
2047790.0069613.4394313.45224819.43054
21151651.5435832.8408022.67515650.629883
3581256.648692.7629242.77943152.87201
161615517.472911.51344911.513449517.4729
33233101.9540943.7537563.60552389.131714
\n", "
" ], "text/plain": [ " I(+) SIGI(+) SIGI(-) I(-)\n", "H K L \n", "20 4 7 790.00696 13.43943 13.45224 819.43054\n", "21 15 16 51.543583 2.840802 2.675156 50.629883\n", "35 8 12 56.64869 2.762924 2.779431 52.87201\n", "16 16 15 517.4729 11.513449 11.513449 517.4729\n", "33 23 3 101.954094 3.753756 3.605523 89.131714" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result2.sample(5)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "A variant of the above function is implemented in `rs.algorithms`, and we will use that implementation in the next section for computing merging statistics. This function computes **IMEAN**, **I(+)**, **I(-)**, and associated uncertainties for each unique Miller index." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "result3 = rs.algorithms.merge(hewl)" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
IMEANSIGIMEANI(+)SIGI(+)I(-)SIGI(-)N(+)N(-)
HKL
38104454.038211.617793416.852521.370535469.638613.841894820
17915101.3088152.1820927100.453273.0412343102.216583.13268783231
38784.5714340.92556394.84819561.27504584.26314931.34570032020
11553991.62947.6906474061.347766.3167653916.956368.632346056
27181114.89911.032750710.9997341.439232219.0381321.48280242020
\n", "
" ], "text/plain": [ " IMEAN SIGIMEAN I(+) SIGI(+) I(-) SIGI(-) N(+) \\\n", "H K L \n", "38 10 4 454.0382 11.617793 416.8525 21.370535 469.6386 13.841894 8 \n", "17 9 15 101.308815 2.1820927 100.45327 3.0412343 102.21658 3.1326878 32 \n", "38 7 8 4.571434 0.9255639 4.8481956 1.2750458 4.2631493 1.3457003 20 \n", "11 5 5 3991.629 47.690647 4061.3477 66.316765 3916.9563 68.63234 60 \n", "27 18 11 14.8991 1.0327507 10.999734 1.4392322 19.038132 1.4828024 20 \n", "\n", " N(-) \n", "H K L \n", "38 10 4 20 \n", "17 9 15 31 \n", "38 7 8 20 \n", "11 5 5 56 \n", "27 18 11 20 " ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result3.sample(5)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Merging with 2-fold Cross-Validation \n", "\n", "To compute correlation coefficients we will repeatedly split our data into half-datasets. We will do this by randomly splitting the data using the image number. These half-datasets will be merged independently and used to determine uncertainties in the correlation coefficients. We will first write a method to randomly split our data, and then we will then write a method that automates the sampling and merging of multiple half-datasets. " ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "def sample_halfdatasets(data):\n", " \"\"\"Randomly split DataSet into two equal halves by BATCH\"\"\"\n", " batch = data.BATCH.unique().to_numpy(dtype=int)\n", " np.random.shuffle(batch)\n", " halfbatch1, halfbatch2 = np.array_split(batch, 2)\n", " half1 = data.loc[data.BATCH.isin(halfbatch1)]\n", " half2 = data.loc[data.BATCH.isin(halfbatch2)]\n", " return half1, half2" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "def merge_dataset(dataset, nsamples):\n", " \"\"\"\n", " Merge DataSet using inverse-variance weighting scheme. This represents the \n", " maximum-likelihood estimator of the mean of the observed intensities assuming \n", " they are independent and normally distributed with the same mean. \n", " \n", " Sample means across half-datasets can be used to compute the merging statistics CC1/2 and CCanom.\n", " \"\"\"\n", " dataset = dataset.copy()\n", " samples = []\n", " for n in range(nsamples):\n", " half1, half2 = sample_halfdatasets(dataset)\n", " mergedhalf1 = rs.algorithms.merge(half1)\n", " mergedhalf2 = rs.algorithms.merge(half2)\n", " result = mergedhalf1.merge(mergedhalf2, left_index=True, right_index=True, suffixes=(1, 2))\n", " result[\"sample\"] = n\n", " samples.append(result)\n", " return rs.concat(samples).sort_index()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Merge HEWL data\n", "\n", "We will now merge the HEWL data, repeatedly sampling across half-datasets in order to assess the distribution of correlation coefficients." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "# This cell takes a few minutes with nsamples=15\n", "merged = merge_dataset(hewl, 15)" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
IMEAN1SIGIMEAN1I(+)1SIGI(+)1I(-)1SIGI(-)1N(+)1N(-)1IMEAN2SIGIMEAN2I(+)2SIGI(+)2I(-)2SIGI(-)2N(+)2N(-)2sample
HKL
004657.9981743.882294657.9981743.882294657.9981743.88229444662.4020425.35386662.4020425.353859662.4020425.35385912120
4645.7966343.884262645.7966343.884266645.7966343.88426644666.474525.35348666.474525.35348666.474525.3534812121
4662.019825.348951662.019825.348951662.019825.3489511212659.1399543.907776659.1399543.907776659.1399543.907776442
4660.469433.18889660.469433.18889660.469433.1888977661.94629.271538661.94629.271538661.94629.271538993
4666.302324.359608666.302324.35961666.302324.359611313639.6683350.654987639.6683350.654987639.6683350.654987334
............................................................
4510230.1702187.351112430.1702187.3511124NaNNaN1015.3181423.968718815.3181423.9687188NaNNaN309
27.6694446.62976077.6694446.6297607NaNNaN1022.894684.108478522.894684.1084785NaNNaN3010
211.6650725.279424711.6650725.2794247NaNNaN2024.1198834.65663424.1198834.656634NaNNaN2011
27.6694446.62976077.6694446.6297607NaNNaN1022.894684.108478522.894684.1084785NaNNaN3012
217.9619124.28813117.9619124.288131NaNNaN3020.0649196.018067420.0649196.0180674NaNNaN1013
\n", "

187364 rows × 17 columns

\n", "
" ], "text/plain": [ " IMEAN1 SIGIMEAN1 I(+)1 SIGI(+)1 I(-)1 SIGI(-)1 N(+)1 \\\n", "H K L \n", "0 0 4 657.99817 43.882294 657.99817 43.882294 657.99817 43.882294 4 \n", " 4 645.79663 43.884262 645.79663 43.884266 645.79663 43.884266 4 \n", " 4 662.0198 25.348951 662.0198 25.348951 662.0198 25.348951 12 \n", " 4 660.4694 33.18889 660.4694 33.18889 660.4694 33.18889 7 \n", " 4 666.3023 24.359608 666.3023 24.35961 666.3023 24.35961 13 \n", "... ... ... ... ... ... ... ... \n", "45 10 2 30.170218 7.3511124 30.170218 7.3511124 NaN NaN 1 \n", " 2 7.669444 6.6297607 7.669444 6.6297607 NaN NaN 1 \n", " 2 11.665072 5.2794247 11.665072 5.2794247 NaN NaN 2 \n", " 2 7.669444 6.6297607 7.669444 6.6297607 NaN NaN 1 \n", " 2 17.961912 4.288131 17.961912 4.288131 NaN NaN 3 \n", "\n", " N(-)1 IMEAN2 SIGIMEAN2 I(+)2 SIGI(+)2 I(-)2 SIGI(-)2 \\\n", "H K L \n", "0 0 4 4 662.40204 25.35386 662.40204 25.353859 662.40204 25.353859 \n", " 4 4 666.4745 25.35348 666.4745 25.35348 666.4745 25.35348 \n", " 4 12 659.13995 43.907776 659.13995 43.907776 659.13995 43.907776 \n", " 4 7 661.946 29.271538 661.946 29.271538 661.946 29.271538 \n", " 4 13 639.66833 50.654987 639.66833 50.654987 639.66833 50.654987 \n", "... ... ... ... ... ... ... ... \n", "45 10 2 0 15.318142 3.9687188 15.318142 3.9687188 NaN NaN \n", " 2 0 22.89468 4.1084785 22.89468 4.1084785 NaN NaN \n", " 2 0 24.119883 4.656634 24.119883 4.656634 NaN NaN \n", " 2 0 22.89468 4.1084785 22.89468 4.1084785 NaN NaN \n", " 2 0 20.064919 6.0180674 20.064919 6.0180674 NaN NaN \n", "\n", " N(+)2 N(-)2 sample \n", "H K L \n", "0 0 4 12 12 0 \n", " 4 12 12 1 \n", " 4 4 4 2 \n", " 4 9 9 3 \n", " 4 3 3 4 \n", "... ... ... ... \n", "45 10 2 3 0 9 \n", " 2 3 0 10 \n", " 2 2 0 11 \n", " 2 3 0 12 \n", " 2 1 0 13 \n", "\n", "[187364 rows x 17 columns]" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "merged" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Compute $CC_{1/2}$ and $CC_{anom}$ \n", "\n", "We will first assign each reflection to a resolution bin and then we will compute the correlation coefficients" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [], "source": [ "merged, labels = merged.assign_resolution_bins(bins=15)" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [], "source": [ "groupby1 = merged.groupby([\"sample\", \"bin\"])[[\"IMEAN1\", \"IMEAN2\"]]\n", "pearson1 = groupby1.corr(method=\"pearson\").unstack().loc[:, (\"IMEAN1\", \"IMEAN2\")]\n", "pearson1.name = \"Pearson\"\n", "spearman1 = groupby1.corr(method=\"spearman\").unstack().loc[:, (\"IMEAN1\", \"IMEAN2\")]\n", "spearman1.name = \"Spearman\"\n", "results1 = rs.concat([pearson1, spearman1], axis=1)\n", "results1 = results1.groupby(\"bin\").agg([\"mean\", \"std\"])" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PearsonSpearman
meanstdmeanstd
bin
00.9973340.0003160.9988950.000063
10.9979580.0003600.9993630.000066
20.9989710.0003060.9996780.000023
30.9992970.0001020.9996580.000024
40.9993680.0001470.9996870.000022
\n", "
" ], "text/plain": [ " Pearson Spearman \n", " mean std mean std\n", "bin \n", "0 0.997334 0.000316 0.998895 0.000063\n", "1 0.997958 0.000360 0.999363 0.000066\n", "2 0.998971 0.000306 0.999678 0.000023\n", "3 0.999297 0.000102 0.999658 0.000024\n", "4 0.999368 0.000147 0.999687 0.000022" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "results1.head()" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(8, 4))\n", "plt.errorbar(results1.index, results1[(\"Pearson\", \"mean\")], \n", " yerr=results1[(\"Pearson\", \"std\")], \n", " color=\"#1b9e77\", \n", " label=r\"$CC_{1/2}$ (Pearson)\")\n", "plt.errorbar(results1.index, results1[(\"Spearman\", \"mean\")], \n", " yerr=results1[(\"Spearman\", \"std\")], \n", " color=\"#d95f02\", \n", " label=r\"$CC_{1/2}$ (Spearman)\")\n", "plt.xticks(results1.index, labels, rotation=45, ha='right', rotation_mode='anchor')\n", "plt.ylabel(\"Correlation Coefficient\")\n", "plt.xlabel(r\"Resolution Bin ($\\AA$)\")\n", "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n", "plt.grid(axis=\"y\", linestyle='--')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It is important to note the scale on the y-axis -- this dataset is edge-limited, and as such the $CC_{1/2}$ is very high across all resolution bins. The Spearman CC appears higher across all resolution bins except at high resolution, and overall has a lower standard deviation among samples.\n", "This is consistent with our expectation that Spearman CCs are a more robust estimator of correlation than Pearson CCs.\n", "\n", "Let's now repeat this for the anomalous data, computing $CC_{anom}$:" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "merged[\"ANOM1\"] = merged[\"I(+)1\"] - merged[\"I(-)1\"]\n", "merged[\"ANOM2\"] = merged[\"I(+)2\"] - merged[\"I(-)2\"]\n", "\n", "# Similar to CChalf, but we will only look at acentric reflections\n", "groupby2 = merged.acentrics.groupby([\"sample\", \"bin\"])[[\"ANOM1\", \"ANOM2\"]]\n", "pearson2 = groupby2.corr(method=\"pearson\").unstack().loc[:, (\"ANOM1\", \"ANOM2\")]\n", "pearson2.name = \"Pearson\"\n", "spearman2 = groupby2.corr(method=\"spearman\").unstack().loc[:, (\"ANOM1\", \"ANOM2\")]\n", "spearman2.name = \"Spearman\"\n", "results2 = rs.concat([pearson2, spearman2], axis=1)\n", "results2 = results2.groupby(\"bin\").agg([\"mean\", \"std\"])" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PearsonSpearman
meanstdmeanstd
bin
00.4221960.0379980.4979230.029704
10.2333490.0619750.3282630.023917
20.3571820.0517460.4446570.029250
30.4847100.0454540.5433140.031731
40.5790580.0239520.6117390.020141
\n", "
" ], "text/plain": [ " Pearson Spearman \n", " mean std mean std\n", "bin \n", "0 0.422196 0.037998 0.497923 0.029704\n", "1 0.233349 0.061975 0.328263 0.023917\n", "2 0.357182 0.051746 0.444657 0.029250\n", "3 0.484710 0.045454 0.543314 0.031731\n", "4 0.579058 0.023952 0.611739 0.020141" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "results2.head()" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "scrolled": false }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(8, 4))\n", "plt.errorbar(results2.index, results2[(\"Pearson\", \"mean\")], \n", " yerr=results2[(\"Pearson\", \"std\")], \n", " color=\"#1b9e77\", \n", " label=r\"$CC_{anom}$ (Pearson)\")\n", "plt.errorbar(results2.index, results2[(\"Spearman\", \"mean\")],\n", " yerr=results2[(\"Spearman\", \"std\")], \n", " color=\"#d95f02\", \n", " label=r\"$CC_{anom}$ (Spearman)\")\n", "plt.xticks(results2.index, labels, rotation=45, ha='right', rotation_mode='anchor')\n", "plt.ylabel(\"Correlation Coefficient\")\n", "plt.xlabel(r\"Resolution Bin ($\\AA$)\")\n", "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n", "plt.grid(axis=\"y\", linestyle='--')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There is significant anomalous signal across all but the highest resolution bins. The Spearman CCs have smaller error bars than the corresponding Pearson CCs and the Spearman CCs are also higher in most bins. These differences highlight the influence of outlier measurements on the different correlation coefficients." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---\n", "### Summary \n", "\n", "We have used `reciprocalspaceship` to merge a dataset using inverse-variance weights. As part of this analysis, we performed repeated 2-fold cross-validation to compute Pearson and Spearman correlation coefficients and associated uncertainties. This relatively simple procedure to obtain uncertainty estimates for correlation coefficients is seldom done when analyzing merging quality. However, we can see that the standard deviation for computed correlation coefficients can be nearly $\\pm0.1$ for quantities such as $CC_{anom}$. This is worth keeping in mind when analyzing SAD experiments because this dataset is very high quality (edge-limited) when one considers the $CC_{1/2}$. Putting these two correlation coefficients on the same axes emphasizes this point:" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(9, 6))\n", "plt.errorbar(results1.index, results1[(\"Pearson\", \"mean\")], \n", " yerr=results1[(\"Pearson\", \"std\")], \n", " color='#1b9e77', \n", " label=r\"$CC_{1/2}$ (Pearson)\")\n", "plt.errorbar(results1.index, results1[(\"Spearman\", \"mean\")], \n", " yerr=results1[(\"Spearman\", \"std\")], \n", " color='#d95f02', \n", " label=r\"$CC_{1/2}$ (Spearman)\")\n", "plt.errorbar(results2.index, results2[(\"Pearson\", \"mean\")], \n", " yerr=results2[(\"Pearson\", \"std\")], \n", " color='#1b9e77', \n", " linestyle=\"--\", \n", " label=r\"$CC_{anom}$ (Pearson)\")\n", "plt.errorbar(results2.index, results2[(\"Spearman\", \"mean\")], \n", " yerr=results2[(\"Spearman\", \"std\")], \n", " color='#d95f02', \n", " linestyle=\"--\", \n", " label=r\"$CC_{anom}$ (Spearman)\")\n", "plt.xticks(results1.index, labels, rotation=45, ha='right', rotation_mode='anchor')\n", "plt.ylabel(r\"Correlation Coefficient\")\n", "plt.xlabel(r\"Resolution Bin ($\\AA$)\")\n", "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n", "plt.grid(axis=\"y\", linestyle='--')\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Even a simple change to the procedure for computing merging statistics, such as substituting a Spearman correlation coefficient for a Pearson one, can alter the apparent quality of a dataset. By lowering the barrier to implementing new analyses, we hope that `reciprocalspaceship` can encourage the development of more robust indicators of crystallographic data quality." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.5" } }, "nbformat": 4, "nbformat_minor": 4 }