Cancer Genetics and Cytogenetics 134 (2002) 168–171
0165-4608/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved.
PII: S0165-4608(01)00630-6
Short communication
Heterogeneous gene distribution reflects human genome complexity
as detected at the cytogenetic level
Antonio Musio
a,
*, Tullio Mariani
b
, Paolo Vezzoni
a
, Annalisa Frattini
a
a
Istituto di Tecnologie Biomediche, C.N.R., Via Fratelli Cervi, 93, 20090 Segrate (MI), Italy
b
Istituto di Biofisica, C.N.R., Via Alfieri, 1, 56010 Ghezzano (PI), Italy
15 August 2001; received in revised form 18 October 2001; accepted 19 October 2001
Abstract
Human chromosomes are heterogeneous in structure and function and this is the base for the specific band-
ing patterns produced by various chromosome staining techniques. The Human Genome Data Base as of
January 2001 was searched for genes mapped to individual chromosomal bands to study the different aspects
of human genome organization as they appear at the cytogenetic level of resolution. Genes are unequally dis-
tributed both on human chromosomes and chromosome bands. Among more than 5000 genes mapped at in-
dividual bands, 81% were located in G-negative bands, which correspond to half of the human genome. The
main practical value of having a dense genetic physical map of genes is to accelerate the discovery by posi-
tional candidate cloning of human disease genes. Gene content agrees with H3 family isochores and with
GC-rich flavors. Interestingly, two G-positive bands, namely 2p chromosome bandings is 12 and 7q35, con-
tain a high number of genes. The finding of heterogeneity in gene content suggests that chromosome banding
is not only due to differences in gene content. © 2002 Elsevier Science Inc. All rights reserved.
1. Introduction
Human chromosomes can be identified cytogenetically
by their relative size and by a distinctive banding pattern
produced by various staining techniques. The most common
methods are G, R, Q, T, and C banding. The banding pat-
terns produced by these different techniques are related to
each other. In fact, the Q banding pattern, produced by treat-
ment with fluorochromes specific for AT-rich DNA regions
is similar to the G banding pattern, produced by staining
with Giemsa after proteolytic digestion. As a whole, G and
Q banding produce a pattern complementary to R banding.
Finally, T banding identifies the most intensely staining R
bands by employing a more severe heat treatment. T bands
correspond to GC-richest R bands [1]. The powerful tech-
niques of chromosomal banding have given the first indica-
tions that the human genome is indeed organized in discrete
regions and that functional characteristics are not randomly
distributed. According to these observations a refinement of
chromosome bands has been proposed. The human genome
may be considered a mosaic of isochores ranging from the
GC-poorest, L1 and L2, to the GC-richest families, H1, H2,
and H3 [2]. A further different approach partitioned the G-
negative bands into four “flavors” depending on base com-
position and Alu repeat density [3].
In this work, we investigated the chromosome distribu-
tion of the genes mapped at individual chromosomal band
reported in the Human Genome Data Base. Although it has
long been assumed that gene distribution is heterogeneous
and that G-positive bands have a lower gene content com-
pared to G-negative bands, this conclusion is relatively old
and was based on relatively few genes for which fine local-
ization was available. Our present paper updates this analy-
sis by examining over 5000 genes whose precise location
has been clearly established and points out some of the fac-
ets of human band behavior.
2. Materials and methods: data base searching
The data present in the Human Genome Data Base [4] as
of January 2001 was manually searched for genes mapped
to individual chromosomal bands. We have not taken into
account the division into sub-bands, so that our analysis has
been performed at a 400-band resolution.
In addition to G-positive and G-negative bands, the hu-
man genome can be considered a mosaic of isochores and
flavors. Four isochore classes can be distinguished: L1
ϩ
L2
* Corresponding author. Tel.:
ϩ
39-02-2642-2636; fax:
ϩ
39-02-2642-
2660.
E-mail address
: musio@itba.mi.cnr.it (A. Musio).