Machine-based methods

Rashid, M., Sana, S., Raghava, G.R Support vector machine-based method for predicting sub-cellular localization of mycobacterial proteins using evolutionary information and motifs. BMC Bioinformatics 2007, 8,337. 

Garg, A., Bhasin, M., Raghava, G.RS. Support vector machine-based method for subceUular localization of human proteins using amino acid compositions, then order, and similarity search. 

K. Suykens, A tutorial on support vector machine-based methods for classification problems in chemometrics. Anal. Chim. Acta, 2010, 665, 129-145. 

Although direct methods for small and medium size systems require more CPU time than disk based methods, this is in many cases irrelevant. For the user the determining factor is the time from submitting the calculation to the results being available. Over the years the speed of CPUs has increased much faster than the speed of data transfer to and from disk. Many modem machines have quite slow data transfer to disk compared to CPU speed. Measured by the elapsed wall clock time, disk based HF methods are often the slowest in delivering the results, despite the fact that they require the least CPU time. 

It is therefore not surprising that the interest in PyMS as a typing tool diminished at the turn of the twenty-first century and hence why taxonomists have turned to MS-based methods that use soft ionization methods such as electrospray ionization (ESI-MS) and matrix-assisted laser desorption ionization (MALDI MS). These methods generate information-rich spectra of metabolites and proteins, and because the molecular ion is seen, the potential for biomarker discovery is being realized. The analyses of ESI-MS and MALDI-MS data will still need chemometric methods, and it is hoped that researchers in these areas can look back and learn from the many PyMS studies where machine learning was absolutely necessary to turn the complex pyrolysis MS data into knowledge of bacterial identities. 

Fig. 4. Application of bioinformatics tools to 2D-DIGE data analysis. Proteome data consisting of the normalized spot intensity values are exported from the image analysis software and their correlation with clinicopathological data examined. Using informatics tools including clustering algorithms and machine-learning methods, a novel cancer classification based on proteome data is established, and key proteomic features and proteins corresponding to biomarker candidates are identified.

Unlike solution based methods in which the sample is dissolved in an aqueous medium that usually includes some small percentage of mineral acid, in GDMS the sample is not a homogeneous liquid, but a solid, and can therefore take a number of forms (e.g., machined hollow cathodes, compacted disks, and dried solution residue). In the next section, cathode/anode geometry is discussed in the section presented here, an equally important but often overlooked issue, the physical form of the sample, is discussed. 

Procedure of creation of the heat machine based on periodic circulation of hydrogen and increase in the efficiency its operation demands the detailed information on methods of calculation equilibrium P-C-T (pressure - concentration - temperature) of characteristics, thermodynamic, thermalphysic (factors of specific heat conductivity X and heat transfers a depending on temperature and pressure) and kinetic properties of hydrides. Approach to designing HHP as to an individual kind of HHM can be broken on three part [1] ... 

Simplistic and heuristic similarity-based approaches can hardly produce as good predictive models as modern statistical and machine learning methods that are able to assess quantitatively biological or physicochemical properties. QSAR-based virtual screening consists of direct assessment of activity values (numerical or binary) of all compounds in the database followed by selection of hits possessing desirable activity. Mathematical methods used for models preparation can be subdivided into classification and regression approaches. The former decide whether a given compound is active, whereas the latter numerically evaluate the activity values. Classification approaches that assess probability of decisions are called probabilistic. 

---