PubMed Health⌕ Search

Biomedical subjects

C Kulikowski

Publications and source records attributed to C Kulikowski.

4 recordsLinked to original sources
PubMed1994

Automated sequencing of amino acid spin systems in proteins using multidimensional HCC(CO)NH-TOCSY spectroscopy and constraint propagation methods from artificial intelligence.

We have developed an automated approach for determining the sequential order of amino acid spin systems in small proteins. A key step in this procedure is the analysis of multidimensional HCC(CO)NH-TOCSY spectra that provide connections from the aliphatic resonances of residue i to the amide resonances of residue i + 1. These data, combined with information about the amino acid spin systems, provide sufficient constraints to assign most proton and nitrogen resonances of small proteins. Constraint propagation methods progressively narrow the set of possible assignments of amino acid spin systems to sequence-specific positions in the process of NMR data analysis. The constraint satisfaction paradigm provides a framework in which the necessary constraint-based reasoning can be expressed, while an object-oriented representation structures and facilitates the extensive list processing and indexing involved in matching. A prototype expert system, AUTOASSIGN, provides correct and nearly complete resonance assignments with one real and 31 simulated 3D NMR data sets for a 72-amino acid domain, derived from the Protein A of Staphylococcus aureus, and with 31 simulated NMR data sets for the 50-amino acid human type-alpha transforming growth factor.

Amino Acid Sequence↗
PubMed1993

Knowledge-based generation of machine learning experiments: learning with DNA crystallography data.

Though it has been possible in the past to learn to predict DNA hydration patterns from crystallographic data, there is ambiguity in the choice of training data (both in terms of the relevant set of cases and the features needed to represent them), which limits the usefulness of standard learning techniques. Thus, we have developed a knowledge-based system to generate machine learning experiments for inducing DNA hydration pattern classifiers. The system takes as input (1) a set of classified training examples described by a large set of attributes and (2) information about a set of learning experiments that have already been run. It outputs a new learning experiment, namely a (not necessarily proper) subset of the input examples represented by a new set of features. Domain specific and domain independent knowledge is used to suggest subsets of training examples from suspected subpopulations, transform attributes in the training data or generate new ones, and choose interesting ways to substitute one experiment's set of attributes with another. Automatic hydration pattern predictors are of both theoretical and practical interest to DNA crystallographers, because they can speed up a labor intensive process, and because the extracted rules add to the knowledge of what determines DNA hydration.

Artificial Intelligence↗