Algorithms for modeling structural changes in human chromosomes

Abstract

Human cytogenetics is the study of chromosomes (typically at mitotic metaphase). The study of chromosomes has recently become integrated with molecular biology and genomics. Thus, it is an important part of genetics education. However, it is time consuming to train students and clinical technologists to recognize patterns of G-banded human chromosomes because of the dynamic nature of G-band resolutions in different metaphase spreads. Moreover, there are limited resources to obtain the images of abnormal chromosomes. We present in this paper an advanced version of computer based interactive tutorial program capable of simulating chromosome abnormalities, altering chromosome shapes, and manipulating G-band resolutions for human cytogenetic seduction. By simulating chromosomes using digital image processing and pattern recognition, the versatile software, together with various strategies such as website links and dialogs, will provide students with a virtual learning environment for self-practicing and testing, thus transforming the traditionally dry and ineffective approach into an exciting and efficient learning process.

Introduction

Human chromosome modeling has recently been introduced as an effective tool for cytogenetics education [1]. Due to the flexibility of the chromosome models, teaching materials can be greatly enriched to provide new features for student practice. These new features include (1) generation of various metaphase spreads with different chromosome distributions, (2) changeable chromosome shapes and sizes, and (3) changeable G-banding resolution levels.

Whereas karyotyping is an important exercise in general genetics education, the clinical application of karyotyping is to detect and analyze abnormal chromosomes related to diseases. Modeling of abnormal human chromosomes will generate a wide range of different types of abnormal chromosomes which will enrich the resources for advanced cytogenetics training.

Chromosome abnormalities are often due to errors in cell divisions, namely, mitosis for somatic cell production and meiosis for gamete formation. Two major types of chromosome abnormalities are (1) changes in chromosome number resulting in euploid series (involving chromosome sets) and aneuploid series (involving individual chromosomes) and (2) changes in chromosome structure. Non-disjunctions of chromosomes during anaphase are the common cause of abnormal chromosome numbers, whereas chromosome breakages may result in a number of structure changes such as deletion, duplication, inversion, and translocation.

Numerical chromosome abnormalities in the euploid series are fetal in early development. However, not all eneuploid conditions are fatal. Individuals with trisomy 21 (Down syndrome), for example, may live relatively normal lives. Trisomy 18 (Edwards syndrome) and trisomy 13 (Patau syndrome) are more serious [2]. Sex chromosome numbers, however, are more flexible due to the inactivation of X chromosomes and the lack of vital genes on the Y chromosome. Thus, individuals with different sex chromosome numbers (e.g., XXX, XO, XXY, and XYY) may survive. Changes in chromosomes numbers, however, do not generate new chromosome features, and there is no need for specific modeling.

Structural changes, on the other hand, entail endless possibilities of generating novel features which make identification difficult and challenging. To enhance the ability of identifying various types of structural changes, constant practice on chromosomes with different structural abnormalities is a key for success. However, in most laboratories, students do not always have an access to a variety of metaphase spreads with different types of structural changes for practice. To increase the teaching materials for learning chromosome abnormalities, we describe in this paper the use of new algorithms for advanced dynamic chromosome modeling. This new interactive computer program is useful for modeling different structural changes in G-banded chromosomes, many of which are associated with mental and physical illnesses, including cancers [2].