RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT

OLFACTORY BULB

Statistical Methods for Understanding Olfactory Codes in Glomerular Images

Benjamin Auffarth, Agust´ın Gutierrez–Galvez and Santiago Marco

Intelligent Signal Processing group, Department of Electronics, University of Barcelona

Artiﬁcial Olfaction group, Institute for Bioengineering of Catalonia (IBEC), C/Baldiri Reixac 13, 08028 Barcelona, Spain

Keywords:

Olfactory coding, Olfactory bulb, Odorants, Glomeruli, Property–activity relationship, Classiﬁcation, Non-

parametric statistics.

Abstract:

The relationship between physicochemical properties of odor molecules and perceived odor quality is arguably

one of the most important issues in olfaction and the rules governing this relationship remain unknown. Any

given odor molecule will stimulate more than one type of receptor in the nose, perhaps hundreds, and this

stimulation reﬂects itself in the neural code of the olfactory nervous system. We present a method to investigate

neural coding at the glomerular level of the olfactory bulb, the ﬁrst relay for olfactory processing in the brain.

Our results give insights into localization of coding sites, relevance of odorant properties for information

processing, and the size of coding zones.

1 INTRODUCTION

Animals are able to recognize a large number of dif-

ferent odors (Axel, 1995) and this is crucial in social

interaction, feeding, and mating. This discriminatory

performance is due to a series of information process-

ing steps at several levels of the olfactory system, be-

ginning from graded afﬁnity of olfactory receptors to

different odors. Depending on the animal, there are

several hundred olfactory receptor types, in particular

about 1000 types in the rat (Buck and Axel, 1991), the

transaction principles of which is presently not well

understood. Apparently, there are no molecular fea-

tures of the odorant that directly determine perceptive

quality (Sell, 2006).

It has been experimentally found (Malnic et al.,

1999) that each receptor type responds to a broad

range of odorants and each odorant evokes responses

from many different receptor types. The odotope the-

ory (Mori and Shepherd, 1994) is the prevalent view

on olfactory transduction and proposes that each ol-

factory receptor detects a combination of structural

molecular features, although it is not clear which

these features are. These combinations of features

are called odotopes in analogy with epitopes, the anti-

genic determinant of the immune system. Each odor-

ant molecule contains many different properties and

the information about the odorant would then be en-

coded by the combined responses of many types of

receptors, each of which recognizes a speciﬁc subset.

Axons from olfactory neurons are bundled in neu-

ropil structures in the olfactory bulb, called glomeruli,

in a way that each glomerulus receives axons just

from one type of receptor(Bozza et al., 2002). It is

well–established that there is a systematic spatial cod-

ing of chemical properties in glomerular activations

(Johnson et al., 1998; Johnson and Leon, 2007) in

the way that odorants with different chemical struc-

ture and shape generate distinct patterns of glomeru-

lar activation. It has also been found that in the rat

olfactory bulb certain properties correlate with activa-

tion in certain zones (Uchida et al., 2000; Johnson and

Leon, 2000; Mori et al., 2006; Johnson et al., 2007).

In this paper, we present a method to analyze sev-

eral aspects of property–activation relationships in rat

glomerular coding of odorants. The questions we in-

vestigate are: which odorant properties are coded and

where, what is the size of the coding zones, and how

relevant are individual odorant properties to the en-

coding. The last question should also give us infor-

mation about relevance of odorant properties to olfac-

tory processing and thereby their contribution to per-

ception of odor quality.

Our techniques consisted of a nonparametric sta-

Auffarth B., Gutierrez–Galvez A. and Marco S. (2010).

RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT OLFACTORY BULB - Statistical Methods for Understanding Olfactory Codes in

Glomerular Images.

In Proceedings of the Third International Conference on Bio-inspired Systems and Signal Processing, pages 37-44

DOI: 10.5220/0002697200370044

 SciTePress

(a) Undecane (b) 2,3,5,6-

tetramethylpyrazine

trimethylcyclohexanone

(d) Anatomical Lo-

cations

Figure 1: Glomerular activity in response to three odorants and anatomical locations in the olfactory bulb. In the response

maps, red stands for high activation, blue for low activation. It can be seen that odorants can be distinguished – to some extend

– by the spatial activation of glomeruli. 1(a) shows responses to the odorant undecane, which is an alkane (compare 2(a)).

Pixel intensities in 1(b) express activation in response to 2,3,5,6-tetramethylpyrazine, which is an aromatic (compare ﬁgure

2(c)). In 1(c) responses to 3,3,5-trimethylcyclohexanone (CAS number 873-94-9) can be seen. This last odorant is a ketone

(compare 2(e)). The rightmost subﬁgure, 1(d), denominates locations in anatomical terms for the glomerular layer of the rat

olfactory bulb. This was adapted from (Johnson et al., 1999).

tistical test with bootstrapping, and support–vector

machine classiﬁcation. Information of these tech-

niques should reveal relevance of odorant properties,

and localization and size of coding zones.

We will explain ﬁrst the data set, then in the meth-

ods section we explain how we approached the ques-

tions, before we look at results, discuss them, and

draw conclusions.

2 DATA SET

In this study we used a set of images of glomerular

responses in the rat olfactory bulb (Johnson et al.,

2006) provided by the group around Michael Leon

and Brett Johnson at the University of California,

Irvine. The images were taken by 2–deoxyglucose

autoradiography and covered the entire lamina. They

took measurements at systematic angle increments

around equally spaced coronal bulb sections. Their

technique has the advantages to work with unanes-

thetized, freely respiring animals and gives the ability

to analyze the entire glomerular layer. On the other

hand, it does not record temporal dynamics of the ol-

factory response and it is impossible to compare re-

sponses in the same animal.

Each of these images corresponds to glomerular

responses to one particular odorant. Examples from

this data set can be seen in ﬁgure 1, which shows

glomerular responses to two odorants.

We mean–centered all pixels to have maps that

show activation at each pixel relative to its overall

pattern, and normalized deviations to standard unit to

compensate for differences in absolute pixel intensi-

ties. We started with 472 maps, some of which rep-

resenting responses to identical odorants in different

concentrations. We took means over concentrations

and discarded a few activity maps for which we did

not have the information of the odorant. This pre–

processing left us with 308 maps corresponding to

distinct and known odorants.

There were missing values in some images. In

the bottom-left of ventral-centered charts they were

caused by loss of tissue on the knife during cryosec-

tioning. Missing values in the central-right parts of

the image were principally due to loss of tissue during

removal of the bulbs from the skull using microdis-

secting scissors. Pixels that had missing values in any

of the images were ignored in the analysis, which left

us with 1834 pixels.

After pre–processing, the data set was a matrix

M ∈ R

308,1834

of activations. Each of the 1834 points

(also: pixels) of this matrix, p ∈ R

308

, represented re-

sponses of the same glomerulus to different odorants.

For all activation maps that remain after pre–

processing we had the information concerning which

odorant they correspond to and additional descrip-

tive information also provided by the Leon Lab. De-

scriptors, about 200 in total, included physicochemi-

cal odorant properties as well as perceptual properties

ascribed to the sensed odor. Properties were of con-

tinuous and binary type. Continuous properties in-

clude molecular length, height, and weight. To give

some examples of binary properties, binary proper-

ties concerned cyclization (whether an odorant is al-

icyclic, aromatic, polycyclic, or heterocyclic), bond

saturation (whether an odorant is alkene, alkane, or

alkyne), and functional groups (whether an odorant is

BIOSIGNALS 2010 - International Conference on Bio-inspired Systems and Signal Processing

ester or lactone, amine, carboxylic acid, contains sul-

fur, contains halogen, is a ketone, alcohol or phenol).

Perceptual properties are all binary and included ﬂa-

vors such as sweet, camphoraceous, ﬂoral, and minty.

For some properties there were many associated activ-

ity maps, for some very few. We had to discard many

because of insufﬁcient representation in the data.

3 METHODS

3.1 Localization of Coding Zones

One of our goals in this research is to determine the

activation loci for each property. This is to say, ﬁnd-

ing the bulbar zones that encode for each property if

such a zone exists.

For each point in the maps and each odorant or

perceptual property, we tested statistically whether

the given pixel shows signiﬁcant differences with re-

spect to the property. For binary properties we com-

pared activations on images, where a property was

given, with activations on images, where property was

not given using a statistical test.

For some properties, we had only very few images

that corresponded to them. To account for statistical

variations in these distributions we used a bootstrap

(Efron, 1982) procedure to estimate p–values of the

statistical test.

The application of the bootstrap to test and de-

rive conﬁdence intervals and p–values was introduced

by (Felsenstein, 1985). Statistical analysis is repeat-

edly applied to subpopulations of the same size, gen-

erated by sampling from the original population with

replacement. Bootstrap methods can be used for hy-

pothesis tests, calculating conﬁdence intervals and re-

gression analysis.

The Wilcoxon ranked–sum test (also called

Mann–Whitney U test) assesses whether two sam-

ples come from the same distribution (null hypoth-

esis). It is analogous to applying the student t–test

on the data after ranking over the combined samples.

It has the advantage of not assuming normality and

of more robustness with respect to the t–test and al-

lows the two samples to be of arbitrary (unequal) size.

The assumptions of the Wilcoxon rank–sum test are

independence of the two samples and independence

of observations within samples, and that the data are

comparable. These assumptions are true for our data

set. Our two samples are the activations given the bi-

nary property and the activations not given the binary

property. The two samples are independent from each

other and activations within samples are also indepen-

dent. They represent the same space, that of activa-

tions, hence they are comparable.

At each iteration we randomly sampled from the

two distributions with replacement before applying

the Wilcoxon rank–sum test. The resulting distribu-

tion of p–values was log–normal and we took the me-

dians of p–valuesas bootstrap statistics and used these

median p–values for subsequent analysis (Limpert

et al., 2001). As estimation of the bootstrap error, we

took the interquartile range of the sampled p–values.

We found that there was a very high and very signif-

icant Pearson correlation between error and p–values

(ρ = 0.77, p = 0.001). About 94% of points below

signiﬁcance level 0.05 had an associated error below

0.1. We only took these points into account (in order

to exclude spurious results).

This method avoids the need to make assumptions

about the shape of the distribution, such as normal-

ity, and uses instead the observed distributions of our

data.

We say that points are coding for a (binary) prop-

erty if the null hypothesis could be rejected at the 5%

signiﬁcance level.

For continuous properties the procedure was more

involved. We discretized properties by grouping their

values into bins, taking bin numbers as ﬁrst guess

from Sturges’ formula (cf. (Wand, 1997)) then ad-

justing so that in each bin there were at least roughly

5% of activation maps. We then applied the procedure

with boostrap and Wilcoxon rank–sum test for differ-

ences between activations in response to property val-

ues in a particular bin versus activations in response

to values out of bin, i. e. testing whether points corre-

sponded to different ranges of the distribution of the

chemical property.

3.1.1 Size of Coding Zones

We investigated the sizes of the zones that coded for

properties. In order to determine the size of coding

zones for a property, we deﬁned the size as the num-

ber of points that were found to be signiﬁcantly dif-

ferent with respect to the property.

Skewed distributions for some properties could

have an impact on how many points are found to be

signiﬁcantly related to a property. By the statistical

test it should be much more difﬁcult for very skewed

distributions to pass the signiﬁcance threshold. In this

paper, for size of coding zones, we take only into ac-

count 13 binary molecular properties, where at least

4 images were available. It is important to note that

for these properties, data availability (odorants corre-

sponding to presented odorant properties) and size of

coding zone show no signiﬁcant Pearson correlation

(ρ = 0.33, p = 0.27).

RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT OLFACTORY BULB - Statistical Methods for

Understanding Olfactory Codes in Glomerular Images

3.2 Classiﬁcation

The method presented in this subsection is based

on the idea that classiﬁcation performance between

glomerular activation and odorant features can give

information about this structure–activation relation-

ship. Speciﬁcally we take the classiﬁcation per-

formance to compare relevances of properties to

glomerular coding. We performed classiﬁcation us-

ing a linear support vector machine (SVM) from

glomerular activations as input vector and each prop-

erty (present vs. not present) as target. In each of 10

iterations we randomly sampled half of the activation

maps as training set and took the other half as test.

We distinguished between two experimental con-

ditions:

1. best points – classiﬁcation using most representa-

tive points, and

2. random baseline – classiﬁcation using randomly

sampled points.

For the ﬁrst experimental condition, for each

property, we ranked points by their signiﬁ-

cance with respect to the property (p–values

from Wilcoxon rank–sum test) and then clas-

siﬁed taking the best n points, with n ∈ N =

[1, 5, 10, 15, 20, 25, 30, 45, 50, 60, 70, 80, 90, 100, 110,

120, 130, 140, 150, 200, 300, 400, 500, 600, 700, 800,

900, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700,

1834]. As a random baseline, for each property,

we took the same intervals from N, but randomly

sampled points. We averaged over 250 random

subsamples of points for each interval.

The scarceness of data for some properties

brought about problems. We found that two com-

monly used SVM implementations, Svmlight and lib-

svm, are lacking robustness to tackle our problem.

Because of this we used an in–house SVM classiﬁer

implemented in Matlab. We used the area under the

ROC curve (AUC) as performance criterion. It has

the advantage to be unbiased by skewed class distri-

butions, which are a particular problem in our data

set. An example of such an experimental run for the

aromatic property is shown in ﬁgure 3.

4 RESULTS

4.1 Localization of Coding Zones

Figure 2 shows loci of coding zones for the 13 molec-

ular properties.

In ﬁgure 2(a), colors indicate where signiﬁcant

differences with respect to alkane were found. This

subﬁgure is to illustrate results from the statistical de-

termination of coding zones for a single property.

For the other chemical properties displayed in ﬁg-

ure 2 we grouped properties into molecular bonds, cy-

clization, and functional groups. We created a fac-

torial code so that the color code accounts for all

combinations of coding for properties. For n prop-

erties, numbers from 0 to n− 1 were assigned to each

property. For each point, a binary vector expresses

whether a property was found to signiﬁcant or not.

The ith position in this vector stands for property i.

Each vector represents a subset of all possible combi-

nations b

prop

∈ { 0, 1}

. Each subset was assigned its

distinct color.

Colors in ﬁgures in 2 show all combinations of

properties that were encountered. To give an exam-

ple, in 2(b) there are seven kinds of zones that mark

codes for different combinations of properties alkane,

alkene, and alkyne. Zones 1, 2, and 4 code for ex-

clusively one of these properties. Zones 3 encodes

alkane and alkene, zone 5 alkane and alkyne, zone 6

alkene and alkyne, and ﬁnally zone 7 codes for all of

the three properties.

Cyclization properties, especially alicyclic, have

a moderate but highly signiﬁcant Pearson correlation

(ρ = 0.33, p = 2.05e− 9 between alicyclic and poly-

cyclic, ρ = 0.36, p = 6.72e−11 between alicyclic and

heterocyclic, and ρ = 0.21, p = 1.95e − 04 between

aromatic and heterocyclic). As can be seen in 2(c) and

2(d), properties aromatic and heterocyclic and prop-

erties alicyclic and polycyclic, respectively, project to

very similar bulbar regions. Functional groups did not

have a high covariance, however there are many prop-

erties (6). To provide clearer ﬁgures, we split coding

zones of both, cyclization and functional groups into

two ﬁgures.

4.1.1 Size of Coding Zones

Table 1 shows size of coding zones as estimated.

From the table it can be seen that aromatic is broadly

coded by glomerular activations. Nearly 60% of

points were found to show differences signiﬁcant at

the 5% level. Alkane covers the second biggest area

with about 40% of points. Carboxylic acid and ketone

are coded by about a third of all points. Coding zones

for properties alkene, alicyclic, and heterocyclic ex-

tend to between about 20 and 30%. For ester+lactone,

alkyne, and alcohol+phenol coding zones we mea-

sured between 10 and 16 percent of total. Properties

polycyclic, sulfur–containing compound and amine

recruit the smallest zones of compared properties with

about 7%, 4%, and 0.6%, respectively.

BIOSIGNALS 2010 - International Conference on Bio-inspired Systems and Signal Processing

(a) Alkane

(b) Molecular Bonds

(d) Cyclization II

(e) Functional Groups I

(f) Functional Groups II

Figure 2: Localization of Molecular Properties: Maps with coding zones for various odorant properties. Loci of the 13

binary properties, grouped into basic dimensions molecular bonds, cyclization, and functional groups. Colors in ﬁgures

serve to distinguish zones, which coded for a speciﬁc combination of several binary properties. For space efﬁciency legends

refer to numbers which are explained in this caption. Factorial maps for all properties of cyclization and functional groups,

respectively, were too crowded and therefore each were broken into two to be better intelligible. 2(a) shows the coding zone

for the alkane property as an demonstration of results of our statistical method of loci determination for a single property.

1 marks the coding zone for alkane. In 2(b), which represents molecular bond properties, the numbers stand for: 1 alkane

2 alkene 3 alkane, alkene 4 alkyne 5 alkane, alkyne 6 alkene, alkyne 7 alkane, alkene, alkyne. 2(c) shows 2 cyclization

properties. The number code is as follows: 1 aromatic 2 alicyclic 3 aromatic and alicyclic. 2(d) shows the other 2 cyclization

properties. The number code is as follows: 1 polycyclic 2 heterocyclic 3 polycyclic and heterocyclic. 2(e) highlights coding

zones for 3 functional group properties. The number code explained: 1 amine 2 ketone 3 amine and ketone 4 alcohol–phenol

5 amine and alcohol–phenol 6 ketone and alcohol–phenol. 2(f) details loci for the other 3 functional group properties. The

numbers: 1 ester+lactone 2 carboxylic acid 3 ester+lactone and carboxylic acid 4 sulfur-containing compound 5 ester+lactone

and sulfur-containing compound 6 carboxylic acid and sulfur-containing compound 7 ester+lactone, carboxylic acid and

sulfur-containing compound. Compare with table 1 where estimations of coding zone size are listed.

4.2 Classiﬁcation

Here we present only results pertaining to 13 molecu-

lar properties, for which at least 4 images were avail-

able. It is important to note that Pearson correlation

between classiﬁcation performance and availability of

data was low and very insigniﬁcant (ρ = 0.2, p = 0.5).

Table 2 ranks properties according to the classi-

ﬁcation performance (AUC) of the linear SVM. The

classiﬁcation performance is indicated in the second

column.

Of the 13 compared properties, sulfur–containing

compound, akyne, alkane, alkene, and amine perform

close to ceiling. Classiﬁcations of carboxylic acid,

aromatic, and ketone also shows good performances.

Polycyclic, ester–lactone, the functional group alco-

hol+phenol, and cyclization properties heterocyclic

and alicyclic gives mediocre performances.

RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT OLFACTORY BULB - Statistical Methods for

Understanding Olfactory Codes in Glomerular Images

Table 1: Sizes of Coding Zones. The table shows for each

property the numbers of points found to be signiﬁcantly cor-

related at 5% signiﬁcance level.

property size of zone

aromatic 1070

alkane 717

carboxylic acid 646

ketone 459

alkene 424

alicyclic 399

heterocyclic 315

ester+lactone 296

alkyne 254

alcohol+phenol 204

polycyclic 125

sulfur-containing compound 76

amine 10

Table 2: Classiﬁcation performance of odorant properties.

The second column shows the maximum classiﬁcation per-

formance (in AUC) that was achieved in baseline or repre-

sentative conditions (whichever was best).

property max performance

sulfur–containing compound 0.99

alkyne 0.99

alkane 0.99

alkene 0.99

amine 0.99

carboxylic acid 0.93

aromatic 0.86

ketone 0.78

polycyclic 0.76

ester+lactone 0.75

alcohol+phenol 0.73

heterocyclic 0.73

alicyclic 0.72

5 DISCUSSION

5.1 Localization and Size of Coding

Zone

We conﬁrmed that for certain properties coding sites

are clustered in zones. This could be the result of an

optimization for local processing of feature combina-

tions (Laughlin and Sejnowski, 2003). Figure 2 illus-

trates some of the properties pertaining to important

coding dimensions as proposed by Johnson and Leon

(Johnson and Leon, 2007). So far, our results of lo-

calization of properties seem to be in accordance with

Figure 3: Classiﬁcation Performance for Aromatic Prop-

erty. The ordinate stands for the number of sampled points

for classiﬁcation. Curves depict performance in AUC with

taking most representative points (solid line), randomly

sampled points (dashed line), and the difference between

the two (dash–dotted line).

literature.

As for the size of coding zone, it should be cau-

tioned that results should be interpreted more quali-

tatively than quantitatively. The thresholding of p–

values at certain signiﬁcance values (here 5%) brings

with it that effects of concentration and relevance

cannot be completely separated from the size, how-

ever presented results can serve to group properties

roughly by their coding zones.

There seem to be very few results in the literature

on sizes of coding zones. Our results present a ﬁrst

step into the direction of quantifying different aspects

of coding at the glomerular level. It seems there are

huge differences with respect of size of coding zone.

There are properties which recruit bigger zones and

properties that recruit smaller zones. This could indi-

cate that some properties are more speciﬁc with re-

spect to the bulbar zone. Some more implications

are discussed below together with the classiﬁcation

results.

5.2 Classiﬁcation

We take classiﬁcation performance of a molecular

property to be indicative of its impact on early ol-

factory coding and — by implication — on percep-

tion. The logic behind is that properties that greatly

change activations at the olfactory bulb level are eas-

ier to classify. Knowing the relevance of molecular

properties could provide insight into early coding of

chemical information and provide vital clues for dis-

cerning which properties are functional in determin-

ing the degree of interaction between an odorant re-

ceptor and odorant molecules.

For some properties, the performance curve from

BIOSIGNALS 2010 - International Conference on Bio-inspired Systems and Signal Processing

representative points is below baseline at some inter-

vals (compare ﬁgure 3). We think this to be because of

the imperfect deﬁnition of most representative points.

We deﬁne relevance as the best classiﬁcation perfor-

mance from either most representative points or ran-

dom baseline whichever was higher.

The performance of the classiﬁcation of amine

found an early peak at 200 points, stayed at high lev-

els until 600 points before leveling offdrastically. The

early peak could be explained by a very small area

corresponding to amine (compare table 1) and in fact

activation maps for amine looked very different from

each other. Taking more points would not provide

more information, rather noise, to the classiﬁcation.

We also think that the support vector machine had dif-

ﬁculties because of only very few data samples corre-

sponding to amine (4 of 308 maps).

Activations are very distinct with respect to

whether an odorant contained an sulfur–containing

functional group or not. Bond saturation indicative of

the reactionarity of compounds seems also to affect

coding very strongly, as we can see in the high per-

formance of alkyne, alkane, and alkene. Carboxylic

acid another functional group and aromatic, a cycliza-

tion property, still seemed to be quite important. So

far, our results conﬁrm that cyclization, bond satura-

tion, and some functional groups are very important.

This is in line with Johnson and Leon (Johnson and

Leon, 2007), who proposed as important dimensions

of molecular properties cyclization, carbon numbers,

bond saturation, branching, functional groups, and

substitution position.

Our results can also be seen to partly conﬁrm

Yoshida and Mori (Yoshida and Mori, 2007) who

proposed 14 primary odorant categories which could

serve to enhance category–proﬁle selectivity. These

properties were sulﬁdes, alcohols, methoxypyrazines,

6–carbon and 9–carbon green–odor compounds, alde-

hydes, ketones, isothiocyanates, terpene hydrocar-

bons, esters, terpene alcohols, alkylamines, acids, lac-

tones, and phenol and its derivatives. We found that

as for the properties included in this study, sulﬁdes,

alcohols–phenol, ketones, ester–lactone, amines, per-

formance was quite good, however our results indi-

cate that other properties such as whether odorants

contained a carboxylic–acid group or their bond satu-

ration could also be very important.

6 CONCLUSIONS

The glomerular level of the olfactory bulb is the ﬁrst

relay for olfactory processing in the brain. The infor-

mation from the glomerular level is factored in sec-

ondary structures with cortical downstream to give

the perception of odor. It has been conﬁrmed that

glomerular activations determine to some degree per-

ceptual qualities of odorants (Cleland et al., 2002).

There has been lot of investigation about which prop-

erties have most impact on perception or representa-

tion in the olfactory bulb, but we are not aware of any

large–scale study to compare many different proper-

ties across a large data set. Our study is a ﬁrst step

into this direction.

We present a method to investigate coding at the

glomerular level of the olfactory bulb and present re-

sults. Our method consisted of the application of the

Wilcoxon rank–sum test within a bootstrap wrapper

and the application of a support vector machine clas-

siﬁcation procedure.

By our statistical procedure we found coding

zones in clustered glomerulifor several properties and

the exact locations of coding zones. By extension

we estimated the size of coding zones and found that

investigated properties differed largely. The proper-

ties for which we found the smallest coding zones are

amine and sulfur–containingcompound, with roughly

0.5% and 4.1% of recruited area.

We then classiﬁed molecular properties by acti-

vation of glomerular activations in order to estimate

relevance of properties. Our classiﬁcation results in-

dicate that there are some properties that affect odor

coding on the olfactory bulb level very strongly. Most

relevant properties we found to be alkyne, alkane,

alkene, and amine. From our study, it could be de-

rived the prediction that these properties have a very

strong impact on perception (at least in rats).

Larger coding zones could mean that properties

are broadly sensed by a range of olfactory receptors.

In turn, it can be conjectured that properties which

have a small coding zone in the olfactory bulb might

have a more direct correspondence to olfactory recep-

tor tuning. It could be hypothesized that properties

with small coding zones could be more directly re-

lated to the proposed odotopes, especially so, prop-

erties that have high relevance to coding. There are

other factors that inﬂuence size of coding zones, such

as lateral connections between glomeruli that compli-

cate matters, however – putting lateral connections

aside – from the results in tables 1 and 2, amine,

sulfur–containing compound, and alkyne could be

candidates for odotopes.

ACKNOWLEDGEMENTS

The authors thank Miquel Tarzan for implementing

a support vector machine that was robust enough to

RELEVANCE AND LOCI OF ODORANT FEATURES IN THE RAT OLFACTORY BULB - Statistical Methods for

Understanding Olfactory Codes in Glomerular Images

make the classiﬁcation study possible. One of the au-

thors, B.A., is supported by a grant from the federal

state government of Catalonia (formaci´o de personal

investigador, FI).

REFERENCES

Axel, R. (1995). The molecular logic of smell. Sci Am,

273(4):154–9.

Bozza, T., Feinstein, P., Zheng, C., and Mombaerts, P.

(2002). Odorant Receptor Expression Deﬁnes Func-

tional Units in the Mouse Olfactory System. Journal

of Neuroscience, 22(8):3033–43.

Buck, L. and Axel, R. (1991). A novel multigene family

may encode odorant receptors: a molecular basis for

odor recognition. Cell, 65(1):175–187.

Cleland, T., Morse, A., Yue, E., and Linster, C. (2002). Be-

havioral models of odor similarity. Behavioral neuro-

science, 116(2):222–231.

Efron, B. (1982). The Jackknife, the Bootstrap and other re-

sampling plans. In CBMS-NSF Regional Conference

Series in Applied Mathematics, Philadelphia: Society

for Industrial and Applied Mathematics (SIAM), 1982.

Felsenstein, J. (1985). Conﬁdence limits on phylogenies: an

approach using the bootstrap. Evolution, pages 783–

791.

Johnson, B., Xu, Z., Pancoast, P., Kwok, J., Ong, J.,

and Leon, M. (2006). Differential speciﬁcity in the

glomerular response proﬁles for alicyclic, bicyclic and

heterocyclic odorants. The Journal of comparative

neurology, 499(1):1–16.

Johnson, B. A., Arguello, S., and Leon, M. (2007).

Odorants with multiple oxygen-containing functional

groups and other odorants with high water solu-

bility preferentially activate posterior olfactory bulb

glomeruli. Journal of Comparative Neurology,

502(3):468–482.

Johnson, B. A. and Leon, M. (2000). Odorant molecular

length: One aspect of the olfactory code. Journal of

Comparative Neurology, 426(2):330–338.

Johnson, B. A. and Leon, M. (2007). Chemotopic odorant

coding in a mammalian olfactory system. Journal of

Comparative Neurology, 503:1–34.

Johnson, B. A., Woo, C. C., Hingco, E. E., Pham, K. L., and

Leon, M. (1999). Multidimensional chemotopic re-

sponses to n-aliphatic acid odorants in the rat olfactory

bulb. Journal of Comparative Neurology, 409(4):529–

548.

Johnson, B. A., Woo, C. C., and Leon, M. (1998). Spatial

coding of odorant features in the glomerular layer of

the rat olfactory bulb. Journal of Comparative Neu-

rology, 393(4):457–471.

Laughlin, S. and Sejnowski, T. (2003). Communication in

neuronal networks. Science, 301(5641):1870–1874.

Limpert, E., Stahel, W., and Abbt, M. (2001). Log-normal

distributions across the sciences: keys and clues. Bio-

science, 51(5):341–352.

Malnic, B., Hirono, J., Sato, T., and Buck, L. (1999). com-

binatorial receptor codes for odors. Cell, 96(5):713–

723.

Mori, K. and Shepherd, G. (1994). Emerging principles of

molecular signal processing by mitral/tufted cells in

the olfactory bulb. Semin Cell Biol, 5(1):65–74.

Mori, K., Takahashi, Y., Igarashi, K., and Yamaguchi, M.

(2006). Maps of odorant molecular features in the

mammalian olfactory bulb. Physiological reviews,

86(2):409–433.

Sell, C. (2006). On the unpredictability of odor. Ange-

wandte Chemie International Edition, 45(38):6254–

6261.

Uchida, N., Takahashi, Y., Tanifuji, M., and Mori, K.

(2000). Odor maps in the mammalian olfactory bulb:

domain organization and odorant structural features.

Nature Neuroscience, 3:1035–1043.

Wand, M. P. (1997). Data-based choice of histogram bin

width. The American Statistician, 51(1):59–64.

Yoshida, I. and Mori, K. (2007). Odorant Category Proﬁle

Selectivity of Olfactory Cortex Neurons. Journal of

Neuroscience, 27(34):9105–9114.

BIOSIGNALS 2010 - International Conference on Bio-inspired Systems and Signal Processing