\n",
+ "\n",
+ "### Feature Finding\n",
+ "\n",
+ "features finding is method for identifying all peaks belonging to a\n",
+ "chemical compound. Feature finding involves the following steps:\n",
+ "\n",
+ "1. **Extension** where we collect all data points we think belong to\n",
+ " the peptide.\n",
+ "2. **Refinement** where we remove peaks that we think do not belong to\n",
+ " the peptide.\n",
+ "3. **Fit an optimal model** to the isolated peaks.\n",
+ "\n",
+ "The above steps are iterative; we repeat these steps until no\n",
+ "improvement can be made to the model.\n",
+ "\n",
+ "OpenMS contains a number of feature finding algorithms.\n",
+ "\n",
+ " Note
\n",
+ "\n",
+ "The term peak picking is ambiguous as it is also used for features\n",
+ "detection (i.e., 3D peak finding).\n",
+ "\n",
+ "
Note
\n",
+ "\n",
+ "While all isotopes are created by adding one or more neutrons to the\n",
+ "nucleus, this leads to different observed masses due to the [mass\n",
+ "defect](https://en.wikipedia.org/wiki/Nuclear_binding_energy#Mass_defect),\n",
+ "which describes the difference between the mass of an atom and the mass\n",
+ "of its constituent particles. For example, the mass difference between\n",
+ "`^{12}C` and `^{13}C` is slightly different than the mass difference\n",
+ "between `^{14}N` and `^{15}N`, even though both only differ by a neutron\n",
+ "from their monoisotopic element:\n",
+ "\n",
+ "``` code\n",
+ "edb = oms.ElementDB()\n",
+ "isotopes = edb.getElement(\"C\").getIsotopeDistribution().getContainer()\n",
+ "carbon_isotope_difference = isotopes[1].getMZ() - isotopes[0].getMZ()\n",
+ "isotopes = edb.getElement(\"N\").getIsotopeDistribution().getContainer()\n",
+ "nitrogen_isotope_difference = isotopes[1].getMZ() - isotopes[0].getMZ()\n",
+ "\n",
+ "print(\"Mass difference between 12C and 13C:\", carbon_isotope_difference)\n",
+ "print(\"Mass difference between 14N and N15:\", nitrogen_isotope_difference)\n",
+ "print(\n",
+ " \"Relative deviation:\",\n",
+ " 100\n",
+ " * (carbon_isotope_difference - nitrogen_isotope_difference)\n",
+ " / carbon_isotope_difference,\n",
+ " \"%\",\n",
+ ")\n",
+ "```\n",
+ "\n",
+ "``` output\n",
+ "Mass difference between 12C and 13C: 1.003355\n",
+ "Mass difference between 14N and 15N: 0.997035\n",
+ "Relative deviation: 0.6298867300208343 %\n",
+ "```\n",
+ "\n",
+ "
Note
\n",
+ "\n",
+ "Protein Ids are linked to peptide Ids by a common identifier (e.g., a\n",
+ "unique string of time and date of the search). The Identifier can be set\n",
+ "using the\n",
+ "[ProteinIdentification.setIdentifier](https://pyopenms.readthedocs.io/en/latest/apidocs/_autosummary/pyopenms/pyopenms.ProteinIdentification.setIdentifier.html)\n",
+ "method in\n",
+ "[ProteinIdentification](https://pyopenms.readthedocs.io/en/latest/apidocs/_autosummary/pyopenms/pyopenms.ProteinIdentification.html)\n",
+ "and\n",
+ "[PeptideIdentification](https://pyopenms.readthedocs.io/en/latest/apidocs/_autosummary/pyopenms/pyopenms.PeptideIdentification.html).\n",
+ "Similarly\n",
+ "[ProteinIdentification.getIdentifier](https://pyopenms.readthedocs.io/en/latest/apidocs/_autosummary/pyopenms/pyopenms.ProteinIdentification.getIdentifier.html)\n",
+ "can be used to check the link between them. With the link one can\n",
+ "retrieve search meta data (which is stored at the protein level) for\n",
+ "individual peptide Ids.\n",
+ "\n",
+ "``` code\n",
+ "```\n",
+ "\n",
+ "> import pyopenms as oms\n",
+ ">\n",
+ "> protein_id = oms.ProteinIdentification() peptide_id =\n",
+ "> oms.PeptideIdentification()\n",
+ ">\n",
+ "> \\# Sets the Identifier protein_id.setIdentifier(\"IdentificationRun1\")\n",
+ "> peptide_id.setIdentifier(\"IdentificationRun1\")\n",
+ ">\n",
+ "> \\# Prints the Identifier print(\"Protein Identifier -\",\n",
+ "> protein_id.getIdentifier()) print(\"Peptide Identifier -\",\n",
+ "> peptide_id.getIdentifier())\n",
+ "\n",
+ "``` output\n",
+ "Protein Identifier - IdentificationRun1\n",
+ "Peptide Identifier - IdentificationRun1\n",
+ "```\n",
+ "\n",
+ "
Note
\n",
+ "\n",
+ "This is an advanced section that dives deep into the\n",
+ "[mzML](https://pyopenms.readthedocs.io/en/latest/user_guide/glossary.html#term-mzML)\n",
+ "format and we will investigate the file format in greater detail. The\n",
+ "intricacies of the\n",
+ "[mzML](https://pyopenms.readthedocs.io/en/latest/user_guide/glossary.html#term-mzML)\n",
+ "file format are all handled by pyOpenMS internally and this section is\n",
+ "only intended for the interested reader\n",
+ "\n",
+ "
Note
\n",
+ "\n",
+ "In the original publication, an E-value is calculated based on the score\n",
+ "distribution p(x), which is derived from a frequency histogram of PSMs\n",
+ "per HyperScore bin, denoted as f(x). The total number of PSMs is\n",
+ "represented by N. The formula for calculating the score distribution is:\n",
+ "$p(x)=\\\\frac{f(x)}{N}$\n",
+ "\n",
+ "For a discrete stochastic score probability distribution p(x), the\n",
+ "so-called survival function represents the probability of having a\n",
+ "greater value than x by random matches in a database. The formula for\n",
+ "the survival function is:\n",
+ "\n",
+ "$s(x)=P(X>x)=\\\\sum{X>xp(x)}$\n",
+ "\n",
+ "To estimate the number of PSMs expected to have scores of x or better,\n",
+ "one can calculate an E-value $e(x)=ns(x)$ Here, n represents the number\n",
+ "of sequences.\n",
+ "\n",
+ "By ranking each PSM in the output according to its E-value, the\n",
+ "significance of individual hits are taken into account. This\n",
+ "functionality is currently not implemented in OpenMS.\n",
+ "\n",
+ "