Project name | Postdoctoral School in the field of Agriculture and Veterinary Medicine |
Project applicant | Banat's University of Agricultural Sciences and Veterinary Medicine Timisoara |
Host institution | University of Agricultural Sciences and Veterinary Medicine Cluj-Napoca |
Grant supervisor | Prof. Dr. Cornel CATOI |
Supervisor assistant | Dr. Cosmina CUC |
Grant call | Admission 2010 |
Implementation timeframe | December 2010 - March 2013 |
Subject category (Tematica) | Metaheuristic algorithms for assessing of flora composition (Algoritmi metaeuristici în evaluarea compozitiei floristice) |
Subject advisor | Prof. Dr. Radu E. SESTRAS |
Grant name (Denumire proiect) | Evaluation of the plant action in relation to structure, composition or genotype using meta-heuristics and evolutionary programming (Evaluarea în flora spontana a actiunii fitofarmaceutice în relatie cu structura, compozitia sau genotipul folosind algoritmi metaeuristici si programe evolutive) |
Grant holder | Lorentz JANTSCHI |
Obtaining of quantitative relationships between plant actions (especially with regard to antioxidant activity) and the genotype, chemical composition, and chemical structure of compounds with biological potency from plant parts (leaves, flowers, fruits, etc..)and obtaining of quantitative measures for assessing biodiversity and bioconservation potential as support for use in decision problems. |
Month | Activity |
01 | Collecting of the information available in the literature at international level |
02 | |
03 | Collecting of the information available in the literature at national and regional (Transylvania region) level |
04 | |
05 | Creating a database: defining the data types and storing manner |
06 | |
07 | Creating a database: establishing of the relationships among information |
08 | |
09 | Design of the queries |
10 | |
11 | Using of the database; seeking for data |
12 | |
13 | Runs of 'training versus test' experiments |
14 | |
15 | Crossover analysis |
16 | |
17 | Intensive measures analysis |
18 | |
19 | Extensive measures analysis |
20 | |
21 | Relating intensive measures with experimental data |
22 | |
23 | Relating extensive measures with experimental data |
24 | |
25 | Capitalization of the knowledge |
26 | |
27 |
Month | Result summary |
01 |
Information in the literature were collected and analyzed. Two works were selected for further study:
|
02 |
Available databases were recorded and studied. Two databases were selected for further analysis:
|
03 |
National and regional range of information were subject of study. English written information were considered. Three sources of interest were selected for further study:
|
04 | Were analyzed the data type of the information which defines the structure of the chemical compounds, the data type for the composition of the mixtures with biological potency, and the data type for the origin and phylogeny of the biological material. Executive summary (in Romanian): În ceea ce priveste structura, informatia este complexa, definind relatii "many-to-many" la nivel topologic si coordonate relative la nivel topografic. Informatia utila din analiza structurii însa nu are acelasi nivel de complexitate, fiind necesar pentru caracterizarea structurii o valoare numerica sau cel mult un sir de valori numerice care sa dea o expresie a contributiei structurii în manifestarea activitatii. Stocarea în schimb a doar acestei valori reduce complexitatea initiala, motiv pentru care s-a ales pastrarea în baza de date a structurii native, însa într-o forma procesabila catre informatia dorita. S-a definit un tip "text" pentru informatia de structura, care sa contina graful molecular în model tridimensional. Compozitia amestecurilor este definita de doua elemente - compus si respectiv pondere (concentratie, fractie molara, etc.) în amestec, asa încât pe lânga informatia stocata cu privire la structura s-a ales sa se stocheze informatia legata de denumire/identificare compus (tip "text") si pondere (tip "numeric"). Informatia de activitate antioxidanta la rândul ei este definita de doua elemente - denumire (ce caracterizeaza si procedura experimentala de determinare, tip "text") si valoare (tip "numeric"). Informatia ce defineste genotipul este o informatie eminamente de clasificare optându-se astfel pentru varianta de stocare tip "text". |
05 |
Three dimensional structure of the chemical compounds has a major implication for the biological activity. The complex type of the 3D structure of the chemical compounds were taken into a deeper study. Storing as the chemical structure as text file was done.
A table with four fields were created for storing the chemical compounds: Id (identifier), CID (PubChem ID), Names (text), Structure (3D, in full), Structure (3D, hydrogen depleted).
Image: |
06 |
Relation between chemical structure and biological activity were considered. A study relating the potency of converting solar energy into chemical energy by the chlorophyls were conducted. 3D models of the cholorophyls structures were obtained and were used to relate with solar energy conversion efficiency.
Image: Manuscript version of the paper: Chlorophylls - natural solar cells Published paper: Chlorophylls - natural solar cells Authors: Lorentz JÄNTSCHI, Sorana D. BOLBOACA, Mugur C. BALAN, Radu E. SESTRAS Acknowledgments: |
07 |
A series of plants from the opposite case, whithout chlorophyls were taken into study: algae from algele Prototheca genus.From NucCore database were downloaded a number of 15 nucleotide sequences for the species: blaschkeae, cutis, moriformis, stagnora, ulmea, and wickerhamii - the last one as complete genome. An analysis of the gene sequences were conducted using literals alignment (see image below).
|
08 |
A study regarding the use of the 2X2 contingency (observed vs. model) and their linkage measures were conducted (see image below).
Acknowledgments: |
09 |
A wider analysis regarding the contingency in effect of essential oil extracts (mixture of compounds) from plant species on bacteria species were conducted. The posibility of factorization of the effect were explored.
A paper capitalized later the research results (image below).
Published paper: Distribution fitting 13. Analysis of independent, multiplicative effect of factors. Application to the effect of essential oils extracts from plant species on bacterial species. Application to the factors of antibacterial activity of plant species Authors: Lorentz JÄNTSCHI, Sorana D. BOLBOACA, Mugur C. BALAN, Radu E. SESTRAS Acknowledgments: |
10 |
Storing of the phylogeny for a certain plant can become a 'hard problem' as can be seen from the study conducted on this subject. By using the same Prototheca genus a study regarding its classification were conducted. As the literature shows, exists different classifications (see image below, from resulted paper).
|
11 |
Same problem of plant extracts (or the effect of mixing for chemical compounds) were taken into the analysis in order to reveal a similarity based on chemical composition (derived from plant metabolism).
The analysis were capitalized in a paper (see image below).
Published paper: Distribution fitting 12. Sampling distribution of compounds abundance from plant species measured by instrumentation. Application to plants metabolism classification Authors: Lorentz JÄNTSCHI, Sorana D. BOLBOACA, Radu E. SESTRAS Acknowledgments: |
12 |
By using a recenty feature of Google Chrome scripting language (so called HTML 5.0 standard, released by Google on August 2011, used here in November 2011) a well known old problem time consuming were solved: uploading of the multiple files to a server from 'one click' (or maybe two), not file-by-file, colectively selected and uploaded. It is a very important problem, because when working with chemical compounds, the compounds must bee find in different databases, downloaded locally, checked, optimized, and uploaded to a local database for analysis. And this procedure should be done in one click, but the security reasons existing in the previous versions of the HTML language denied this option, of multiple selection of files to be uploaded.
The image below gives the developed script using this feature (see image below).
|
13 |
A valuable database containing a large number of experimental measurements were found. It is Dr. Duke's Phytochemical and Ethnobotanical Databases (http://www.ars-grin.gov/duke/). This database has been used to extract useful information. A series of steps has been followed in order to access the data in a relational database manner. These steps are:
|
14 |
As continuation of the analysis conducted in the previous month, identification of the groups of data in such (as Duke's) databases were the subject of the investigation. For these particular cases, when a large block of data is available, seeking for linearities is possible. A procedure for seeking these linearities were developed. Two online applications are available as well as their results of analysis of the Duke's database:
|
15 |
Data treatment were taken into study in order to obtain the coefficients for crossover. Following list iterates the steps:
|
16 |
In order to give a true estimate of the crossover when only a part (M < N) of a paired data (of size N) is taken all possible extactions should be made, and the the average result is a true estimate of the cross validation. In order to do this, the algorithm described in the paper [Phillip J. CHASE, 1970. Algorithm 382: Combinations of M out of N Objects [G6]. Communications of the Association for Computing Machinery 13(6):368-368] were used to implement the succesive draws of M elements from a set of N. The procedure has ben further tested on the full (15 pairs of data) and the normalized data (14 pairs of data) from the Duke's database. The results are given in the next image.
|
17 |
Intensive measures for diversity were taken into analysis. Two measures were selected for further analysis: Renyi Entropies family (left in the image below) as being representative for observed diversity and Fisher's alpha (right in the image below) as being reprezentative for estimated diversity [Refs: Rényi, A. 1961. On measures of information and entropy. Proceedings of the 4th Berkeley Symposium on Mathematics, Statistics and Probability, 547-561; Fisher, RA. 1943. Part 3. A theoretical distribution for the apparent abundance of different species. Journal of animal ecology 12:54-58].
|
18 |
Mobility. A series of conclusions has been drawn from the study visit in Germany and Holland, given below:
|
19 |
Among with Fisher's method, other methods of diversity estimation were taken into study, namely:
Published paper: Distribution fitting 16. How many colors are actually in the field? Authors: Lorentz JÄNTSCHI Acknowledgments: |
20 |
The rarefaction method for estimating the diversity from the sample were implemented (see image below).
|
21 |
The same study from previous month were implemented with the result from combinatorics giving the number of rarefied colors. The implemented code is given below:
|
22 |
The use of the entropy measures (Renyi) to compare the genus based on chemical composition were conducted. The results are given below:
|
23 |
The use of the molecular families of deschiptors is a manner to relate the chemical structure with the biological activity. A study regarding the distribution of the correlation coefficients were conducted in order to identify the type of the distribution for the set of descriptors providing an agreement between the chemical structure and the observed property by using molecular descriptors obtained via MDFV methodology [Bolboaca SD, Jäntschi L, 2009. Comparison of QSAR Performances on Carboquinone Derivatives, TheScientificWorldJO 9(10):1148-1166. DOI: 10.1100/tsw.2009.131]. The toxicity measured at different stages of development for the species Arbacia punctulata, Dinophilus gyrociliatus, Sciaenops ocellatus, Opossum shrimp and Ulva fasciata. A number of 24 observed biological activities for a number of 8 compounds served in this investigation [U.S. Geological Survey, Marine Ecotoxicology Research Station, Texas A&M University-Corpus Christi, Center for Coastal Studies. Development of marine sediment toxicity for ordnance compounds and toxicity identification evaluation studies at select naval facilities. http://web.ead.anl.gov/ecorisk/issue/pdf/tox_marine_sed.pdf]. The results shown a partition of distribution functions as below.
Authors: Lorentz JÄNTSCHI, Sorana D. BOLBOACA Acknowledgments: |
24 |
A new methodology relating the chemical structure with the biological activity were designed: SAPF. The result were capitalized in a publication (see image below).
Authors: Radu E. SESTRAS, Lorentz JÄNTSCHI, Sorana D. BOLBOACA Acknowledgments: Authors: Radu E. SESTRAS, Lorentz JÄNTSCHI, Sorana D. BOLBOACA Acknowledgments: |
25 |
A study continuing the research from PhD Thesis in Horticulture (2010) were conducted in order to estimate the moments of evolutions in different selection and survival strategies. The study reveals that the relative moments of evolutions are shaped by a one-parameter degeneration of the log-Pearson type III distribution. The results conducted on a given data sample allowed to extract the parameters of these distributions (see image below).
Authors: Lorentz JÄNTSCHI, Sorana D. BOLBOACA, Radu E. SESTRAS Acknowledgments: |
26 |
Further capitalization of the knowledge from the study conducted in the project were regarding the distribution of the seeds sizes (data from the buyed book describing the species from Transylvania region). A very nice picture of the seeds sizes distribution were obtained (see image below).
Authors: Lorentz JÄNTSCHI, Rodica C. SOBOLU, Sorana D. BOLBOACA Acknowledgments: |
27 |
Further capitalization of the knowledge from the study conducted in the project were regarding the effect of the leverage and of the influential on the quality of the structure-activity relationships.
The study shows that the Di model has the biggest change relative to the initial model (see image below)
Authors: Sorana D. BOLBOACA, Lorentz JÄNTSCHI Acknowledgments: |