Build Packages / Unit tests (push) Skipped
Build Packages / build:windows:nocuda (push) Successful in 11m6s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 10m27s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 10m54s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 9m25s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 10m5s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 11m33s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 11m19s
Build Packages / build:rpm (rocky8) (push) Successful in 12m23s
Build Packages / build:rpm (rocky9) (push) Successful in 13m21s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 12m30s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 11m55s
Build Packages / DIALS test (push) Successful in 13m42s
Build Packages / XDS test (durin plugin) (push) Successful in 9m26s
Build Packages / XDS test (JFJoch plugin) (push) Successful in 6m41s
Build Packages / XDS test (neggia plugin) (push) Successful in 6m12s
Build Packages / Generate python client (push) Successful in 19s
Build Packages / Build documentation (push) Successful in 52s
Build Packages / Create release (push) Skipped
Build Packages / build:viewer-tgz:cpu (push) Successful in 5m29s
Build Packages / build:viewer-tgz:cuda (push) Successful in 6m12s
Build Packages / build:windows:cuda (push) Successful in 18m36s
This is an UNSTABLE release. It includes many experimental features, as well as many AI generated fixes. We recommend using rc.152 for production use. * rugnux: Add `--model model.pdb` - score the merged data against an atomic model and compute initial maps. It reports R-work/R-free (scaling the model to the observed amplitudes with an overall scale, an anisotropic B and a flat bulk solvent - the standard few-parameter model, so a batch of maps stays directly comparable) and writes 2Fo-Fc / Fo-Fc electron-density maps (CCP4) plus a map-coefficient MTZ. The structure itself is not refined; the model is only re-fractionalised into the data cell. * rugnux: The merged reflection output now carries French-Wilson amplitudes (|F| and its sigma) next to the intensities - MTZ `F`/`SIGF`, mmCIF `_refln.F_meas_au`, and the text HKL - computed with the correct centric/acentric Wilson prior and epsilon multiplicity, so a downstream program (e.g. phenix.refine) can refine against amplitudes. The intensity columns are unchanged. * rugnux: R-free test-set flags are now assigned deterministically and consistently across symmetry - a Bijvoet pair I(+)/I(-) is never split between the work and free sets, and the assignment is a reproducible per-hkl hash that depends only on the reflection index, so every dataset of one crystal form gets the same ~5% free set (what a multi-dataset campaign such as PanDDA needs). On small data the fraction is floored so the test set stays large enough for a stable R-free (~500 reflections, capped at 10%); it stays flat at 5% on ordinary data. When a reference MTZ carries a `FreeR_flag` column its test set is imported instead, letting a whole campaign inherit one shared free set. * rugnux: A reference MTZ (`--reference-mtz`) can now fix the space group and cell for rotation data too (previously rejected), without being used to scale - the rotation merge stays self-consistent. When the crystal has an indexing (merohedral) ambiguity - a lattice symmetry higher than its Laue symmetry, e.g. P3/P4/P6/C2 - the reference also resolves it: each candidate reindexing (identity plus the twin-law cosets of the metric symmetry) is scored by its intensity correlation against the reference and the data are re-merged in the best-correlating one. This is a metric-preserving relabelling of hkl (the cell is unchanged) and a no-op for a holohedral crystal such as lysozyme. * rugnux: `--model` validation now aligns the data to the model before scoring - the observed reflections are reindexed into the model's enantiomorph when the two differ only by hand (indistinguishable from merged intensities). A merohedral indexing ambiguity is resolved against the reference MTZ when one is given (so a whole campaign shares one indexing convention); only with a model and no reference does validation fall back to fitting each candidate reindexing and keeping the lowest R-free. * rugnux: De-novo symmetry - recover a genuine high-symmetry group whose data are imperfectly scaled. Such a merge's within-orbit chi² lands just past the self-consistency bound (each real symmetry step adds a little systematic scatter), right where a merohedral twin also lands, so the chi² ratio alone cannot separate them. The candidate is now rescued when the extra intensity-proportional systematic error it invokes stays small relative to the confirmed subgroup - a genuine symmetry step gains multiplicity without inflating the merge error model's b, whereas a twin forces non-equivalent reflections together and b balloons. Fixes cubic insulin (I23 instead of I222) with no change to any other crystal in the test battery, including the twins that must stay in their lower symmetry. * Docs: Document the French-Wilson amplitude estimation, R-free flagging, reference-based space-group/ambiguity resolution, and model-based validation/maps in CPU_DATA_ANALYSIS.md. * Frontend: The status-bar pill now shows a progress bar during detector calibration (previously only during measurement), and the calibration state and its button are labelled "Calibration"/"CALIBRATE" (the internal `Pedestal` state name is unchanged for back-compatibility).Reviewed-on: #69 Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
210 lines
6.3 KiB
C++
210 lines
6.3 KiB
C++
// Copyright 2019 Global Phasing Ltd.
|
|
|
|
// Calculation of atomic form factors approximated by a sum of Gaussians.
|
|
// Tables with numerical coefficients are in it92.hpp and c4322.hpp.
|
|
|
|
#ifndef GEMMI_FORMFACT_HPP_
|
|
#define GEMMI_FORMFACT_HPP_
|
|
|
|
#include <cmath> // for exp, sqrt
|
|
#include <cstdint> // for int32_t
|
|
#include <cstring> // for memcpy
|
|
#include <limits> // for numeric_limits
|
|
#include <utility> // for pair
|
|
#include "math.hpp" // for pi()
|
|
|
|
namespace gemmi {
|
|
|
|
// NOTE: the argument x must be between -88 and 88.
|
|
// It is based on expapprox() from
|
|
// https://github.com/jhjourdan/SIMD-math-prims/blob/master/simd_math_prims.h
|
|
// Relative error is below 1e-5.
|
|
inline float unsafe_expapprox(float x) {
|
|
static_assert(std::numeric_limits<float>::is_iec559, "float is not IEEE 754?");
|
|
//static float zero = 0.f; // non-const to disable optimization
|
|
float val = 12102203.1615614f * x + 1065353216.f;
|
|
//val = std::max(val, zero); // check if x < -88.02969
|
|
std::int32_t vali = static_cast<std::int32_t>(val);
|
|
std::int32_t xu1 = vali & 0x7F800000;
|
|
std::int32_t xu2 = (vali & 0x7FFFFF) | 0x3F800000;
|
|
float a, b;
|
|
std::memcpy(&a, &xu1, 4);
|
|
std::memcpy(&b, &xu2, 4);
|
|
return a * (0.509871020f + b * (0.312146713f + b * (0.166617139f + b *
|
|
(-2.190619930e-3f + b * 1.3555747234e-2f))));
|
|
}
|
|
|
|
// precalculated density of an isotropic atom
|
|
template<int N, typename Real>
|
|
struct ExpSum {
|
|
Real a[N], b[N];
|
|
|
|
Real calculate(Real r2) const {
|
|
Real density = 0;
|
|
for (int i = 0; i < N; ++i)
|
|
density += a[i] * std::exp(b[i] * r2);
|
|
return density;
|
|
}
|
|
|
|
std::pair<Real,Real> calculate_with_derivative(Real r) const {
|
|
Real density = 0;
|
|
Real derivative = 0;
|
|
for (int i = 0; i < N; ++i) {
|
|
Real y = a[i] * std::exp(b[i] * (r * r));
|
|
density += y;
|
|
derivative += 2 * b[i] * r * y;
|
|
}
|
|
return std::make_pair(density, derivative);
|
|
}
|
|
};
|
|
|
|
template<int N>
|
|
struct ExpSum<N, float> {
|
|
float a[N], b[N];
|
|
float calculate(float r2) const {
|
|
float density = 0;
|
|
float tmp[N];
|
|
for (int i = 0; i < N; ++i)
|
|
tmp[i] = std::max(b[i] * r2, -88.f);
|
|
for (int i = 0; i < N; ++i)
|
|
density += a[i] * unsafe_expapprox(tmp[i]);
|
|
return density;
|
|
}
|
|
|
|
std::pair<float,float> calculate_with_derivative(float r) const {
|
|
float density = 0;
|
|
float derivative = 0;
|
|
float tmp[N];
|
|
for (int i = 0; i < N; ++i)
|
|
tmp[i] = std::max(b[i] * (r * r), -88.f);
|
|
for (int i = 0; i < N; ++i) {
|
|
float y = a[i] * unsafe_expapprox(tmp[i]);
|
|
density += y;
|
|
derivative += y * b[i] * (2 * r);
|
|
}
|
|
return std::make_pair(density, derivative);
|
|
}
|
|
};
|
|
|
|
// precalculated density of an anisotropic atom
|
|
template<int N, typename Real>
|
|
struct ExpAnisoSum {
|
|
Real a[N];
|
|
SMat33<Real> b[N];
|
|
|
|
Real calculate(const Vec3& r) const {
|
|
Real density = 0;
|
|
for (int i = 0; i < N; ++i)
|
|
density += a[i] * std::exp(b[i].r_u_r(r));
|
|
return density;
|
|
}
|
|
};
|
|
|
|
template<int N>
|
|
struct ExpAnisoSum<N, float> {
|
|
float a[N];
|
|
SMat33<float> b[N];
|
|
|
|
float calculate(const Vec3& r_) const {
|
|
Vec3f r((float)r_.x, (float)r_.y, (float)r_.z);
|
|
float density = 0;
|
|
float tmp[N];
|
|
for (int i = 0; i < N; ++i)
|
|
tmp[i] = std::max(b[i].r_u_r(r), -88.f);
|
|
for (int i = 0; i < N; ++i)
|
|
density += a[i] * unsafe_expapprox(tmp[i]);
|
|
return density;
|
|
}
|
|
};
|
|
|
|
template<typename Real>
|
|
Real pow15(Real x) { return x * std::sqrt(x); }
|
|
|
|
// Gaussian coefficients with functions to calculate sf and density.
|
|
template<int N, int WithC, typename Real>
|
|
struct GaussianCoef {
|
|
using coef_type = Real;
|
|
static const int ncoeffs = N;
|
|
std::array<Real, 2*N+WithC> coefs;
|
|
Real a(int n) const { return coefs[n]; }
|
|
Real b(int n) const { return coefs[N+n]; }
|
|
Real c() const { return WithC ? coefs[2*N] : 0; }
|
|
|
|
void set_coefs(const std::array<Real, 2*N+WithC>& c) { coefs = c; }
|
|
|
|
// argument: (sin(theta)/lambda)^2
|
|
Real calculate_sf(Real stol2) const {
|
|
Real sf = c();
|
|
for (int i = 0; i < N; ++i)
|
|
sf += a(i) * std::exp(-b(i)*stol2);
|
|
return sf;
|
|
}
|
|
|
|
Real calculate_density_iso(Real r2, Real B) const {
|
|
constexpr Real _4pi = Real(4 * pi());
|
|
Real r2pi = Real(r2 * pi());
|
|
Real density = c() * pow15(_4pi / B) * std::exp(-(_4pi / B) * r2pi);
|
|
for (int i = 0; i < N; ++i) {
|
|
Real t = _4pi / (b(i)+B);
|
|
density += a(i) * pow15(t) * std::exp(-t*r2pi);
|
|
}
|
|
return density;
|
|
}
|
|
|
|
// note: addend is considered only if WithC (addend is usually dispersion f')
|
|
ExpSum<N+WithC,Real> precalculate_density_iso(Real B, Real addend=0) const {
|
|
ExpSum<N+WithC,Real> prec;
|
|
constexpr Real _4pi = Real(4 * pi());
|
|
for (int i = 0; i < N; ++i) {
|
|
Real t = _4pi / (b(i)+B);
|
|
prec.a[i] = a(i) * pow15(t);
|
|
prec.b[i] = -t * Real(pi());
|
|
}
|
|
if (WithC) {
|
|
Real t = _4pi / B;
|
|
prec.a[N] = (c() + addend) * pow15(t);
|
|
prec.b[N] = -t * Real(pi());
|
|
}
|
|
return prec;
|
|
}
|
|
|
|
Real calculate_density_aniso(const Vec3& r, const SMat33<float>& U) const {
|
|
constexpr Real pi2 = sq(pi());
|
|
const SMat33<Real> B = U.scaled(8 * pi2);
|
|
Real density = c() * pow15(4 * pi()) / std::sqrt(B.determinant()) *
|
|
std::exp(-4 * pi2 * B.inverse().r_u_r(r));
|
|
for (int i = 0; i < N; ++i) {
|
|
SMat33<Real> Bb = B.added_kI(b(i));
|
|
density += a(i) * pow15(4 * pi()) / std::sqrt(Bb.determinant()) *
|
|
std::exp(-4 * pi2 * Bb.inverse().r_u_r(r));
|
|
}
|
|
return density;
|
|
}
|
|
|
|
ExpAnisoSum<N+WithC,Real> precalculate_density_aniso_b(const SMat33<Real>& B,
|
|
Real addend=0) const {
|
|
constexpr Real m4pi2 = Real(-4 * sq(pi()));
|
|
constexpr Real pow_4pi_15 = (Real) 44.546623974653663; // pow15(4 * pi())
|
|
ExpAnisoSum<N+WithC,Real> prec;
|
|
for (int i = 0; i < N; ++i) {
|
|
SMat33<Real> Bb = B.added_kI(b(i));
|
|
prec.a[i] = a(i) * pow_4pi_15 / std::sqrt(Bb.determinant());
|
|
prec.b[i] = Bb.inverse().scaled(m4pi2);
|
|
}
|
|
if (WithC) {
|
|
prec.a[N] = (c() + addend) * pow_4pi_15 / std::sqrt(B.determinant());
|
|
prec.b[N] = B.inverse().scaled(m4pi2);
|
|
}
|
|
return prec;
|
|
}
|
|
|
|
ExpAnisoSum<N+WithC,Real> precalculate_density_aniso_u(const SMat33<float>& U,
|
|
Real addend=0) const {
|
|
constexpr Real UtoB = 8 * sq(pi());
|
|
return precalculate_density_aniso_b(U.scaled(UtoB), addend);
|
|
}
|
|
};
|
|
|
|
} // namespace gemmi
|
|
#endif
|