Automated classification

B Oliva, PA Bates, E Querol, LX Aviles, MIL Sternberg. An automated classification of the structure of protein loops. I Mol Biol 266 814-830, 1997. 

Of course, classification of images of proteins whose location is already known does not contribute much to our understanding of protein location. We can, however, imagine applying a trained classifier to large collections of images of proteins whose location is not known. Automated classification is a powerful approach, but... 

Huang, K. and Murphy, R.F. (2004a) Boosting accuracy of automated classification of fluorescence microscope images for location proteomics. BMC Bioinformatics 5, 78. 

Tate AR, Majos C, Moreno A, Howe PA, Griffiths JR, Ams C. Automated classification of short echo time in vivo IH brain tnmor spectra a multicenter stndy. Magn. Reson. Med. 2003 49 29-36. 

Fehrenbach, U., Scherer, D., and Parlow, E. (2001) Automated classification of planning objectives for the consideration of climate and air quality in urban and regional planning for the example of the region of Basel/Switzerland, Atmos. Environ. 35(32), 5605-5615. 

Prediction of Drug-Induced PT Toxicity and Injury Mechanisms with an hiPSC-Based Model and Machine Learning Methods The weak points of the HPTC- and hESC-based models described previously (Sections 23.3.2.1 and 23.3.3.1) were the data analysis procedures. In order to improve result classification, the raw data obtained with three batches of HPTC and the 1L6/1L8-based model (Li et al., 2013) were reanalyzed by machine learning (Su et al., 2014). Random forest (RE), support vector machine (SVM), k-NN, and Naive Bayes classifiers were tested. Best results were obtained with the RF classifier and the mean values (three batches of HPTC) ranged between 0.99 and 1.00 with respect to sensitivity, specificity, balanced accuracy, and AUC/ROC (Su et al., 2014). Thus, excellent predictivity could be obtained by combining the lL6/lL8-based model with automated classification by machine learning. 

Mazur, A.L (2013) Spectral pathology automated classification of cytological and histological specimens utilizing infrared micro-spectroscopy, in Chemistry Chemical Biology, Northeastern University, Boston, MA,... 

For homogeneous NDT data and repeatable inspection conditions successful automated interpretation systems can relatively easily be developed. They usually use standard techniques from statistical classification or artificial intelligence. Design of successful automated interpretation systems for heterogeneous data coming form non-repeatable, small volume inspections with little a-priori information about the pieces or constructions to be inspected is far more difficult. This paper presents an approach which can be used to develop such systems. 

In order to make as much data on the structure and its determination available in the databases, approaches for automated data harvesting are being developed. Structure classification schemes, as implemented for example in the SCOP, CATH, andFSSP databases, elucidate the relationship between protein folds and function and shed light on the evolution of protein domains. 

When applied to QSAR studies, the activity of molecule u is calculated simply as the average activity of the K nearest neighbors of molecule u. An optimal K value is selected by the optimization through the classification of a test set of samples or by the leave-one-out cross-validation. Many variations of the kNN method have been proposed in the past, and new and fast algorithms have continued to appear in recent years. The automated variable selection kNN QSAR technique optimizes the selection of descriptors to obtain the best models [20]. 

U.S. Patent and Trademark Office Web Patent Databases. The Patent and Trademark Office (PTO) [72] offers free World Wide Web access, http // www.uspto.gov/main/patents.htm, to a bibliographic patent database that uses the most current patent classification system, this may not match the classification data that appears on the printed patent, and to a full-text patent database that uses the classification data that appear on the printed patent, this may not match the current classification data. The databases start with January 1, 1976, patents. The full text of a patent includes all bibliographical data (e.g., inventor s name, the patent s title, the assignee s name, etc.) and the abstract, full description of the invention, and the claims. All the words in the text of the patent are searchable. If the patent number is known, the patent, regardless of year, can be ordered from the PTO. Automated searching of 1971 to date patents is available at some of the Patent and Trademark Depository Libraries. Prior to 1971 searching can be done at the PTO facilities or at the Patent and Trademark Depository Libraries. Commercial patent search services are also available. 

These examples illustrate the power of proper ANN feature space optimization. In all the examples discussed, the limits of the type of information that could be gleaned from the Salmonella PyMAB spectra were probed. The PD-ANN s automated optimization removed the issue of methodological uncertainty and enabled a focus on questions of Py-MAB-MS spectral information content and its potential use for rapid strain ID. Question Does Py-MAB-MS data support Serovar classification Answer Yes. How about PFGE classification Yes. How about antibiotic resistance profile Answer Perhaps, if one first eliminates stronger contributions to spectral variation and then, by design and grouping, limits the possibilities to only a few classes. 

Because of this, a straightforward quantitative approach would not suffice, even if one could be developed. We need methods to deal with the existence of errors in the training classifications when training instruments to do automated identification. 

MS has recently been used to measure compounds with significant levels of impurities and solubilities below the quantitation limits of other methods. Guo et al.46 described the use of LC/MS for solubility measurements in buffer solutions in a 96-well plate. Fligge et al.47 discussed an automated high-throughput method for classification of compound solubility. They integrated a Tecan robotic system for sample preparation in 384-well plates and fast LC/MS for concentration measurement. This approach is limited by LC/MS throughput. 

Fligge, T.A. and Schuler, A. 2006. Integration of a rapid automated solubility classification into early validation of hits obtained by high throughput screening. J. Pharm. Biomed. Anal. 42 449. 

Immunoaffinity chromatography (IAC), 6 400—402 12 137, 145 Immunoanalyzers, automated, 14 150 Immunoassay(s), 14 135-159. See also Immunoassay- DNA probe hybrid assays Immunoassay methods Immuno(bio)sensors antibody-antigen reaction, 14 136-138 basic technology in, 14 138-140 chemiluminescent, 14 150-151 classification of, 14 140-153 design of, 14 139-140 enzyme, 14 143-148 fluorescence, 14 148-150 highly specific, 14 153 historical perspective on, 14 136 microarrays and, 14 156—157 microfluidics in, 26 968—969 monoclonal versus polyclonal antibodies in, 14 152-153... 

Because of the complexity of integrated processes and the large volume of available data in highly automated plants, classification algorithms are increasingly used nowadays. They are applied to the design of monitoring systems and to reduce the dimension of the data reconciliation problem. 

Part—I has three chapters that exclusively deal with General Aspects of pharmaceutical analysis. Chapter 1 focuses on the pharmaceutical chemicals and their respective purity and management. Critical information with regard to description of the finished product, sampling procedures, bioavailability, identification tests, physical constants and miscellaneous characteristics, such as ash values, loss on drying, clarity and color of solution, specific tests, limit tests of metallic and non-metallic impurities, limits of moisture content, volatile and non-volatile matter and lastly residue on ignition have also been dealt with. Each section provides adequate procedural details supported by ample typical examples from the Official Compendia. Chapter 2 embraces the theory and technique of quantitative analysis with specific emphasis on volumetric analysis, volumetric apparatus, their specifications, standardization and utility. It also includes biomedical analytical chemistry, colorimetric assays, theory and assay of biochemicals, such as urea, bilirubin, cholesterol and enzymatic assays, such as alkaline phosphatase, lactate dehydrogenase, salient features of radioimmunoassay and automated methods of chemical analysis. Chapter 3 provides special emphasis on errors in pharmaceutical analysis and their statistical validation. The first aspect is related to errors in pharmaceutical analysis and embodies classification of errors, accuracy, precision and makes... 

Doppelt-Azeroual O, Delfaud F, Moriaud F et al (2010) Fast and automated functional classification with MED-SuMo an application on purine-binding proteins. Protein Sci 19 847-867... 

Brakoulias A, Jackson RM (2004) Towards a structural classification of phosphate binding sites in protein-nucleotide complexes an automated all-against-all structural comparison using geometric matching. Proteins Struct Funct Bioinformatics 56 250-260... 

---