Research Statement


1. Statement of teaching philosophy

I believe as a biologist and a teacher it is my responsibility to challenge and support my students, in pursuit of three learning goals:

(1) Learn basic statistical conceptsfor biological research;

(2) Cultivate an interest in science and an understanding of how we can measure nature;

(3) Apply statistical tools to some biological phenomena.

Yet, students will have to make their own decisions about biological issues. I want my students to leave my classroom with basic biological skills that enable them to engage with these ideas, making well-considered choices.

I gather students’ initial ideas about the main component of an ecosystem; I also would like to give them the concepts of phylogenyand an organism’s ontogenesis.

On the base of this, I would like to teach the issue of developmental stability and other problems of developmental biology. This is a concept and a theory. I would share methods of testing developmental stability using detection fluctuating, asymmetry, directional asymmetry and antisymmetry. The first type of asymmetry performs environmental reaction on stress and plays a role in environmental studies. Two others refer to the genotypic nature and are very helpful in genetic monitoring.

Thus, my approach is multidisciplinary. It includes the study of ecology, toxicology, statistics and basic biological courses (zoology, botany, evolution). The crucial stage in my course is that of developing experiments to test developmental stability/instability in various plants, insects, or birds. In most biology lab classes, students carry out lab activities by following step-by- step directions printed in the my lab manuals. In this stage students do not experience the joy of discovery, since every step of the experiment, including expected results, is explicitly stated, requiring little creative thought.

In my lab classroom, I use an inquiry-based curriculum, centring on the principle that students should actually do science themselves. Exploring science entails defining a biological problem or question, addressing the problem through observing living creatures’ communities/ecosystems, and interpreting the findings. An inquiry-based curriculum offers many opportunities for active learning, creating an atmosphere in which students must take responsibility for their learning. During one three-week-long segment, my students test the developmental stability of populations of woody plants and evaluate factors that are significant for deviation from normal stability.

Students are immediately immersed in many aspects of science and life; working in small groups, through intensive discussion, they collaboratively design and conduct research, deciding which factors are the most significant and why.  After carrying out their experiments, students write articles in the style of popular magazines in which they interpret their experimental results and translate these ideas into everyday language.

I have two main benchmarks for determining that my objectives for students are met: that students are able to convey biological concepts in everyday language; and that students are able to use basic knowledge and skills as building blocks to tackle more complex problems. More importantly, this may be their final biology course. The students develop the ability to collaborate and to follow directions. These skills are transferable to many aspects of their lives.

Computational biology, as a discipline, is concerned with transforming “data” into “information.” Thus, a statistical education that explains both how to learn from data and how to make inferences from data has an important place in a well-rounded biological education. My objectives as a general biology educator are:

(1) Teach students about statistical tools and how to use them correctly;

(2) Expose students to examples of statistical analysis applicable in routine biological scientific activity; and

(3) Teach students how to communicate statistical results and ideas clearly to a variety of audiences.

These objectives apply to all the courses I teach, with varying degrees of emphasis depending on the level of the course. My teaching philosophy can be summarized by the following beliefs:

▪ Students learn statistical theory best when they see how this material can be applied in real, practical activity.

▪ Students must be able to see past the computational drudgery to the underlying principles. Both oral and written skills in their presentations are important.

▪ Concerning computer packages: students in introductory or service courses in statistics should use a computer package (e.g. JMP) to analyze realistic problems. This frees the students from worrying about the computational monotony often associated with statistics and mathematics and allows them to concentrate on the concepts.

▪ My assignments always include written questions where students must explain their results. Thus, they practice developing clear writing.

Upper-level students must complete term papers, and graduate students are strongly encouraged to make oral presentations about their work to future enhance their speaking skills.




My research work is the testing of developmental stability/instability level. The method I apply is a testing of the fluctuating asymmetry (FA) value in bilaterally symmetrical biological structures. So I define (characterize) my activity as a computational approach to the Systems biology and to the ecology of population.

My work is finalised by my conclusions about: a) the level of FA value and, correspondingly, b) the level of developmental stability (instability) and c) the stressful factors affecting the studied population. The FA method is reliable for studying all terrestrial and water plants with bilateral asymmetry. Therefore, the method is applicable in ecology of population first.

The next courses can be provided in the frame of the University curriculum. They are Developmental Stability, Biostatistics for Ecology, Biological Asymmetry, Computational Ecology.

The main objectives are to use and analyze row data treatment (Excel), the normality of distribution, the types of asymmetry (directional, non-directional and FA, antisymmetry), the criteria of their evaluation.

The statistical method I used is the Two Way ANOVA. The approach I am using is the Geometric morphometry. The softs I apply are PAST, TPS, SAGE, STAISTICA and MorphoJ).

Preface, Terms

The fluctuating asymmetry has been used for more than 30 year to estimate stress influence on an ecosystem or its separate components (the trees, the shrubs, the grass, the amphibians and the insects).

Terms: Fluctuating asymmetry (FA), directional asymmetry (DA), developmental instability (DI), statistically significant data, environmental stress (ES).

Predicted Summaries/outputs

My proposals implicate the students’ activity in the frame of scientific work, and in following activity in the research institutes, in the environmental groups etc. The proposals require a participation as undergraduate BSC students as well postgraduate for MS and PhD degree. As a social activity these proposals are aimed to involve students in researching of wild ecosystem health. This means the evaluation of the developmental stability of living things, for example of plants that have bilaterally symmetrical organs (e.g. leaves, flowers).

The predictable outputs are the publishing of articles, the increasing of students’ study activity, the evaluation of the environmental situation in ecosystems for sustainable development of some regions under stress, and the studying of climatic changes and result of human (anthropogenic) impact. The economical result means a financially appropriate approach for environmental monitoring and management.


The evaluation of developmental stability means a testing of the level of fluctuating asymmetry of bilaterally symmetrical homological traits. The high deviation from the perfect symmetry means a deviation from the normal development of stability. The concept of fluctuating asymmetry assumes the geographical mapping and finally the monitoring for sustainable development of socially industrial complex (fig.1).

Fig.1. On the way to sustainable development

A Student’s Individual Tasks


The task includes the following: students should collect the leaves (from trees, from bushes, or from grass). The students measure the traits of some species, detect the volume of FA with defining of the DI level, and later they consolidate their forces for the result according to the local scientific community’s goals. The species could be any but they must have bilaterally symmetric traits. The plant families which were studied in works of Russian and foreigner searches in alphabet order:

Species and Samples Requirements

They have to be suitable for testing. They have to be common for the chosen area as an available plant for long-term monitoring and they have to include the bilaterally symmetrical traits (the left and the right), for example the petals of a flower or vessels on a leaf. The sample means the number of specimens collected in locality (cite) of one species population. The leaves (flowers) have to be free of the mechanical damage or visible strong asymmetry. It is a very important that the leaves (flowers) be the same size.

Method of the sample collecting

The final idea of this research is to compare some areas with the value of asymmetry of one or more species. Then it can be possible to make a conclusion about the factor causing the deviation in developmental stability.

Tree plants. 10 leaves from one tree should be collected; the total number of the trees is 10; the total number of the leaves (flowers) for one species is one hundred. As a result there should be collected the same amount of the leaves from the control population, e.g. 200 leaves from two populations (experimental and control).

Grass plants. 30 specimens of the leaves (flowers) of one species (one specimen from one individual grass plant) are collected. These leaves should be kept dry in the paper press. If the leaf or flower is small, they should be photographed. Then we save images as a soft database, for example as JPEG.

Environmental Impact, Collaborative Work and Discussions

The stress factors can be the following: the erosion of the soil, the pesticides effect, the climatic factors (the humidity, the regional/seasonal temperature of the air). The factors ranged from low to high.

The different kinds of activity imply the contact with institutes involved in the biodiversity conservation, in the water supply activity, in the climatic service, in the chemical pollution testing(the heavy metals, the pesticides), and in the crops/vegetables quality testing. All the results could be corresponded with the internet community specialists.

The research output is reversed\prepared for the preliminary report and for the final graduate qualifying research work. The tutors share the statistical and the biological concepts, the methodologies, and the techniques, which allow students the successfully participating in the workshops and computer labs. The students will be strongly encouraged to work with other students to collaborate, and to provide the assistance and peer mentoring.

Trait Measuring

There are two methods of measurement: by the ruler and by PC programs like Photoshop or TPS. The ruler has to be made of stainless steel with a well-calibrated scale. Any units can be used (cm, mm, inches, pcs, others). For the evaluation of the measurement error, the specimen traits must be measured three times. All data is saved in Excel folders with the proper legend for each sample (the date, the area/region, the habitat, the environmental factors like pollution, climatic features, height above the sea level and the character of the soil). For undergraduate students: TPS and SAGE software are used. As example of a leaf (Lime tree) and traits for fluctuating asymmetry (FA) detection is shown in the fig.2.


Fig.2. Tilia cordata leaf traits used for testing FA

1) The width of the leaf on the level of the third vein of the primary order; 2) The length between the base of the first and second vein secondary orders on the first vein of the primary order;3) The length between the base of the second and third vein secondary orders on the first vein of the primary order;4) The length between the base of the first and second veins of the main rib;5) The angle between the central main vein (mid rib) and the first bilateral vein



  • Basic formula for FA value evaluation: FA = |R – L|/ (R+L)
  • For every species, some appropriate traits should be choosen. Comparing two or more areas for each FA trait’s datum should be compared with the same trait’s datum. E.g. the FA value of trait number 1 is compared to the FA value of trait number 1. The FA value trait number 2 is compared to the FA value of trait number 2 etc. Only traits with statistically significant data of FA value can be used for the conclusion about the ecological impact of the environmental factor in terms of developmental stability/instability;
  • The FA value 0.05 – 0.07 corresponds to a medium level of any stress. The FA value less than 0.05 corresponds to a low level of the stress. The level more than 0.07 means a high level of the stress with a serious decrease in developmental stability (increase in developmental instability);
  • I propose a new morphogeometric method for testing FA value. These are excellent method approved by many researchers. The method serves as an advanced tool in teaching students in frame of Biostatistics and Computational biology.


Geometric Morphogeometric Method, GM method

The idea of this method is to use the difference between some point’s coordinates of the left and the right sides in homologous organs. In plants the endpoints of veins, the points of the biggest cavity in sinuses or the endings of the lobes, on the leaf plates are commonly used. The advantage of the method is in the use of 2D Cartesian coordinates. The basic idea of the shape analysis is applying the possible configurations and determining the most appropriate configurations for all samples. It employs the method of “least squares”. Numerically, a form is analysed as the deviation in the variance between the points of the average shape (aligned centroid) and the corresponding points of real samples.

The soft program is employed for these purposes, mostly used for shape analysis of fossil bones and other biological objects, such as representatives of class of fish in cladistics. Work carried out with the leaves of alder, showed a variability shape of the leaf blade, depending on location (Pavlinov, 2002;Klingenberg, 1998). In botanical studies shape analysis is used in taxonomy (Banaev, 2005).

The work in the field of geometric morphometrics of FA testing has been used mainly with Drosophila (Klingenberg, 2001-2008). Much less, the method employed in FA testing of plants, including woody plants (Albarra-Lara, 2010; Baranov, 2012). Figure 3 shows an example of image processing with the hibiscus flower in Excel media after positioning of key points coordinates by the TPS program (Rohlf, 2010).


Fig.3. Bilaterally symmetrical landmarks in inflorescences Hibiscus engleri (Malvaceae):

A - E, B - D; cf – axis of symmetry; (b) – A set of landmarks (50 samples). A1, B1, C1, D1, and E1: points formed after rotation point C to the intersection with the axis of symmetry co-incident with the axis OY (worked out in MS Excel)


The distance from the center coordinates to each of the points was found using the Pythagorean Theorem. This distance is useful for calculating the difference between the values of the left and the right homologous landmarks. However, only the rough testing of FA value is possible with Excel.

In recent years a number of software packages (TPS, MorphoJ, PAST, and others, see appendix) were developed to analyze geometric characteristics performing various operations of editing, converting files and statistical treatment of row data.

Among the multivariate statistical methods, the analysis of principal components and discriminate analysis are widely used (Rohlf, 1996; Saunders, 1993).One of the methodological approaches is the Procrustes method of alignment. According to this method, the right and the left points should aligned along with the mirror-reflected landmarks (fig.4).

The Procrustes analysis includes the original and mirrored configurations of a sample combined, and superimposes all of them simultaneously. For averaging consensus the method of least squares is used.



Fig.4. Displaying of the original figure (solid line) on the basis of the Procrustean alignment (on Klingenberg et al, 2002). Dotted line is a mirror-image copy for each half. Dash-dotted line is the symmetrical average figure forming on the base of least squares method (superimposition). Procrustes FA is the difference between the landmarks coordinates of the original figure and symmetric average standard, or consensus


The methods of geometric morphometry have limitations because they are algebraic, not random and not related to descriptive statistic laws. In order to assay the FA two-factor analysis of variance, the values of the XY coordinates of homologous bilaterally symmetrical points are displayed in the tangent space. According to R. Palmer (2003), the coordinates have the same value as the value of the right and the left homologous traits, and contain information about the instability development (Palmer, 1992). The fundamental advance of the morphogeometric approach is the landmarks carrying data of the shape difference of the homologous symmetrical structures. Thus, the value of angles and characters of shape are taken into account. Initially, two-way Procrustes ANOVA was adopted for FA testing of Drosophila metric traits. For plants this analysis of variance is used to digitalise data for leaf plates of oak and flowers from some plant families' representatives.

The SAGE soft (fig.5) was developed for FA analysis the by Marquez in 2006 and is designed to work with the file format TPS. The program is easy for the user. It allows you not only to identify the integral value of FA in the Procrustean analysis, but allows you also to test a statistical difference in the fluctuating asymmetry of individual landmarks.

When compared with trivial two-way ANOVA assay (individual × side) the Procrustes analysis includes more freedom of degree in the (2k + l - 2) times, where: k – is a number pairs of landmarks; l – is a number of single landmarks on the midline. Therefore, the Fisher level of probability is increased, as this criterion testing the null hypothesis is sensitive to the degree of freedom.

The significance of factor "side" is reporting about the presence of DA, while the value of the mean square interactions between "side" and "individual" means a value of fluctuating asymmetry in a sample.

A GIS-based approach allows us to determine the dependence of FA on response to exposure to stressful environmental factors to predict the level of health of the environment in the early stages of the violation ecosystems, including urban areas.

Review of literature sources in the field of stability of plant development shows that the effect of environmental factors has no one side effect on the level of fluctuating asymmetry. Methodically it is seen the complication of procedures identifying FA with the new statistical methods. At the same time the unification of new software package methods allows to detect quickly and effectively variation in asymmetry including changes in FA level.

Fig.5. Left: Tilia cordata leaf blade and landmarks for testing FA with Procrustes ANOVA. Right (SAGE screenshot: Procrustes fit original data and original+ reflected data with aligned points (colourless in each constellation), using as a points of consensus

Fig.6. Quercus robur. Left – 10 pair of landmarks for testing FA. In the middle – the landmarks represented on a Procrustes fit in symmetric matrix and in matrix of asymmetry (right). Black dots show the landmarks after aliment by superimposition. Symmetric matrix possesses more variance due to variety in lobes and sinuses. Matrix of asymmetry reflects variation of landmarks in left and right sides (MorphoJ 1.06)

Methods of the geometric morphometry detecting FA in dependence on the shape allow determining a genetic component of variations in comparison with the environmental components, while defining a digitizing measurement error. An important advantage is the index of integrative FA values testing all of the studied traits, though, namely individual homologous traits indicate the intraindividual/intraorgan variability. Modern GIS technologies offer the promise of long-term biota monitoring, including monitoring the stability of elements of flora, as the main component of ecosystems cities, urban landscapes and protected areas. Thus, monitoring of developmental stability becomes an integral component of bioindication.


Perspectives in GIS

GIS (Geographic Information System) is a computer collection, storage, processing and display of spatially coordinated data that integrate diverse information, coming from different sources based on spatial position, whereby it is possible to compare a variety of environmental factors and conduct a comprehensive geo-ecological assessment area.

Thus, GIS is a necessary and indispensable backbone part of a regional biomonitoring database and in context e-maps reveals the dynamics of changes of environmental quality in the region and ecological prognosis in future are available.

Application of GIS technology allows receiving bioindicating maps integrated assessment of environmental health surveyed area, and with mathematical precision to compare the individual electronic maps, spending zoning on environmental quality in the automatic mode, with varying degrees of accuracy (updates). Plants as the autotrophic base part of the ecosystem are the most important part of the system bioindication and environmental monitoring. The morphogenetic approach involves assessment of stability development (homeostasis) of bilaterally symmetric organisms. The health of natural populations was evaluated by analyzing the fluctuating asymmetry. This approach has been extended in Kaluga, Dubna, and Kaliningrad. Over 50,000 plant samples and animal things was analyzed (Strel'cov, 2005). Evaluation was carried out on the territory of the individual control points (sites, localities), on the basis of which the territory’s health was evaluated.

 The following graphical methods are used:

1) Placing the analysis results on the map as a sample point diagram (circular, radial, histograms, etc.). This widely used method, and the most accurate, because it shows the true values of the points therefore obtaining information.

2) Construction of a mathematical surface distribution the value of the indicators in the study area by interpolation, and obtaining orthogonal projection of the calculated surface map in the form of contour lines.

Point selection is one of the tasks in GIS. For large-scale analysis it is desirable that the points are uniformly coated with study area. However, in practice, points are often arranged irregularly due in the absence of the point areas relevant species. In general, the frequency of points depends on what level of detail expected to get results.

In the GIS technology a computer program Surfer (Surface Mapping System, ver. 6.04, Golden Software Inc.) is used. This program offers a choice of 8 kinds of interpolation. The method of using Inverse Distance to a Power is meant that the impact value at some point is inversely proportional to the distance from it to the nodes. More remote data points and the method will be less affected by a specific node. The input data is a table containing a rectangular the collection point coordinates (X, Y) and the value of the test indicator (Z) at these points. After interpolation of missing values the regular network or matrix were calculated.

On the basis of numerical matrix the surface distribution of values and the orthogonal projection on the earth's surface (map) in isolines was constructed. The resulting maps are informative and clear.

Since the calculated surface is digital matrix, they can be deducted and summarized. Therefore, to create an integrated map average surface area of all studied matrices species are calculated.


My teaching and consolidated research philosophy can be summarized by the following beliefs:

 ▪ Students learn statistical theory best when they see how this material can be applied in real practical activity.

▪ Students must be able to see past the computational drudgery to the underlying principles. Both oral and written skills of students in their presentations are important.

▪ Using computer packages: Students in introductory or service courses in statistics should use a computer package (e.g. JMP) to analyze realistic problems. This frees the students from worrying about the computational drudgery often associated with Statistics and Mathematics and allows them to concentrate on the concepts.


So the synthesis biology and statistics is a target and main body of my proposal. I believe this is a good entrance and a next wide way into scientific activity forthcoming youth.







Agricultural plants are genetically homozygous and convenient for studying developmental stability. What does the development stability of such populations mean? Is there a contradiction between the main principle in the agricultural industry (productivity) and development stability? If there is a contradiction, i.e. stability does not correlate with productivity, what to choose the first or second? Choosing the stability of development, we take nothing more than a step towards organic and biosphere farming and ignore the high productivity of the culture. High productivity is just a property that is skillfully developed in genetically modified plants. So, what to choose productivity for the sake of low stability or quality (high development stability)? The answer is of course quality! 

Fig. 1. The arrangement of 50 landmarks on the wheat leaf blade. First and second LMs are not paired.


The study showed dependancy FA level from inorganic fertilizer doze (the more dose decreases developmental stability), high level of admixture of FA and DA (directional asymmetry).  This was true for wheat. The study to be continued.

Fig. 2. Rye. First principal components (LM – blue colored, consensus – not filled). Strong FA

Fig. 3. PC1. Asymmetric component Mix DA and FA

The study on leaves rye revealed high and pure FA  under low dose fertiliser (N90P90K90).