Partial structure estimation apparatus and method of generating partial structure estimation model

A computer program product comprising at least one non-transitory computer-readable medium including program instructions that, when executed by at least one processor of an information processing apparatus, cause the at least one processor to:

• control an ion source to ionize a sample to generate a number of fragment ions;
• control a mass spectrometry apparatus to perform mass spectrometry on each of the number of fragment ions to generate a mass spectrum including a group of peaks;
• generate a first explanatory variable by performing composition estimation for each peak based on the group of peaks included in the mass spectrum;
• generate a second explanatory variable by performing composition estimation for each peak interval based on the group of peaks;
• and
• estimate a partial structure existing in the sample as an objective variable based on the first explanatory variable and the second explanatory variable.

A method of generating a partial structure estimation model, the method comprising:

• registering, by at least one processor, a plurality of partial structures;
• ionizing, with an ion source, a known compound to generate a number of fragment ions;
• performing, with a mass spectrometry apparatus, mass spectrometry on each of the number of fragment ions to generate a mass spectrum including a group of peaks;
• generating, by at least one processor, a first explanatory variable by performing composition estimation for each peak based on the group of peaks included in the mass spectrum;
• generating, by at least one processor, a second explanatory variable by performing composition estimation for each peak interval based on the group of peaks;
• generating, by at least one processor, an objective variable by judging presence or absence of the partial structure in the known compound for each partial structure which is registered;
• and
• generating, by at least one processor, a partial structure estimation model through machine learning using training data formed from the first explanatory variable, the second explanatory variable, and the objective variable.