Bioinformatics work notes

As reported in the Ensembl 2009 NAR paper canonical transcripts are defined for all genes and for all species in the Ensembl gene sets. "The canonical transcript is defined as either the longest CDS, if the gene has translated transcripts, or the longest cDNA. Should a transcript already regarded as canonical not be selected using the above rules, there is support for storing this information in the Ensembl database."

These objects can be regarded as representative transcripts for the gene and can be fetched with the Perl API method
Bio::EnsEMBL::Gene::canonical_transcript()

Some caution needs to be used when looking at the pseudo-autosomal regions: When looking at genes from the Y PAR, the method will return a transcript with X coordinates. While not really a bug, this might mess up your data if un-noticed. To check and fix this something like the following will work:

Code:

Telomeres form caps on the ends of chromosomes that prevent fusion of chromosomal ends and provide genomic stability.
During gametogenesis, reprogramming of the germ cells leads to elongation of telomeres up to their species-specific maximum.
In normal somatic cells, telomeres are progressively shortened with every cell division. This shortening in normal human cells limits the number of cell divisions. For human cells to proliferate beyond the senescence checkpoint, they need to stabilize telomere length. This is accomplished mainly by reactivation of the telomerase enzyme. Telomerase expression is under the control of many factors. Expression of telomerase can lead to cell immortalization and is activated during tumorigenesis, i.e. cancer.
Male Xq-telomeres are 1100 bp shorter than female Xq-telomeres.
The telomeric repeat found on all human chromosomes is "TTAGGG".

The centromeres and telomeres of the human chromosomes are not defined as region attributes in the Ensembl perl API explicitely, so for checking these regions, one option is to pull them out of the UCSC table browser and define them manually. You can e.g. create an array of hashes with the regions and use them in your script:

#read data (listed below) from a file...
my @data = split("\s");
my %telomere = (
'chrom' => $data[1],
'start' => $data[2],
'end' => $data[3],
);
push(@telomeres, \%telomere);

The list of centromere regions for GRCh37 is:

1       121535434       124535434
2       92326171        95326171
3       90504854        93504854
4       49660117        52660117
5       46405641        49405641
6       58830166        61830166
7       58054331        61054331
8       43838887        46838887
9       47367679        50367679
10      39254935        42254935
11      51644205        54644205
12      34856694        37856694
13      16000000        19000000
14      16000000        19000000
15      17000000        20000000
16      35335801        38335801
17      22263006        25263006
18      15460898        18460898
19      24681782        27681782
20      26369569        29369569
21      11288129        14288129
22      13000000        16000000
X       58632012        61632012
Y       10104553        13104553

The list of telomere regions for GRCh37 is:

1       0               10000
1       249240621       249250621
2       0               10000
2       243189373       243199373
3       0               10000
3       198012430       198022430
4       0               10000
4       191144276       191154276
5       0               10000
5       180905260       180915260
6       0               10000
6       171105067       171115067
7       0               10000
7       159128663       159138663
8       0               10000
8       146354022       146364022
9       0               10000
9       141203431       141213431
10      135524747       135534747
10      0               10000
11      134996516       135006516
11      0               10000
12      0               10000
12      133841895       133851895
13      0               10000
13      115159878       115169878
14      0               10000
14      107339540       107349540
15      0               10000
15      102521392       102531392
16      0               10000
16      90344753        90354753
18      0               10000
18      78067248        78077248
19      0               10000
19      59118983        59128983
20      0               10000
20      63015520        63025520
21      0               10000
21      48119895        48129895
22      0               10000
22      51294566        51304566
X       0               10000
X       155260560       155270560
Y       0               10000
Y       59363566        59373566

Telomeres of chromosome 17 have not been defined for assembly GRCh37. They are short, but do exists nonetheless. An assembly patch will address this.

Sources:

• Rashid-Kolvear et al. 2007
• Moyzis et al. 1988
• Perner et al. 2003

The karyotype (number and set-up of the chromosomes) of a person or any changes in specific regions on human chromosomes are described with a system of numbers and symbols, defined by a group of cytogentic experts as the International System for Human Cytogenetic Nomenclature (ISCN). The system was based on the ideogram definitions by Francke et al. (1981), and was last revised 2005 and 2009.

The visual bands are created by staining techniques and describe regions of similarity in respect to functionality and base compositions (GC content); their lengths are 5-10 MB.

Example ideograms: Human chromosome 1 in different resolutions, from WashU

Numbers used:
The regions are numbered from the centromer outwards in both directions towards the telomeres on the shorter p arm and the longer q arm. The numbers cannot be read in the normal decimal numeric system e.g. 21, but rather 2-1 (region 2 band 1). Counting starts at the centromer as region 1 (or 1-0), to 11 (1-1) to 21 (2-1) to 22 (2-2) etc. Subbands are added in a similar way, eg. 21.1 to 21.2, if the bands are small or only appear at a higher resolution.
There are different levels of resolution that can be used as bands, e.g. 1q32 in the 400-bands resolution can be split up into 1q32.1, 1q32.2, 1q32.3 in the 550-bands resolution (see Figure above as an example for chr 1).

Here is a list of chromosome, band (arm, region, band, subband), genomic start and end position, from the Ensembl database for assembly GRCh37:

Code:

Symbols used:
The following symbols are often used with band numbers when describing changes in the karyogram, eg. of cancer cells.

 , 	Separates chromosome modal number, sex chromosomes, 
        and chromosome abnormalities
 - 	Loss of a chromosome
 ( ) 	Surround structurally altered chromosomes and breakpoints
 + 	Gain of a chromosome
 ++     Multiple signals on one chromosome
 ; 	Separates rearranged chromosomes and breakpoints involving
        more than one chromosome
 / 	Separates cell lines or clones
 // 	Separates recipient and donor cell lines in bone marrow
        transplants
 ~      approximation
 x      multiple copies of chromosomes or regions
 .      Sperates multiple techniques
 amp    amplification
 arr    microarray data
 dim    diminished fluorescence ratio intensity: deletion
 del 	Deletion
 der 	Derivative chromosome (used when only one chromosome from
        a translocation is present, or when one chromosome has two
        or more structural abnormalities)
 dic 	Dicentric chromosome
 dn 	Chromosomal abnormality not inherited from parents (de novo)
 dup 	Duplication of a portion of a chromosome
 enh 	Enhanced fluorescence ratio intensity: duplication
 fra 	Fragile site (usually used with Fragile-X syndrome)
 h 	Heterochromatic region of chromosome
 hlpa   Multiple ligation-dependent probe amplifications
 hmz    Homozygosity
 htz    Heterozygosity
 i 	Isochromosome (both arms of the chromosome are the same)
 ins 	Insertion of a portion of a chromosome
 inv 	Inversion
 .ish 	Precedes karyotype results from fluorescence in situ
        hybridization (FISH) analysis
 mar 	Marker chromosome (unidentifiable piece of chromosome)
 mat 	Maternally derived chromosome rearrangement
 p 	Short arm of a chromosome
 pat 	Paternally derived chromosome rearrangement
 psu dic 	Only one centromere is active (pseudo dicentric)
 q 	Long arm of a chromosome
 r 	Ring chromosome
 t 	Translocation
 ter 	Terminal end of arm (i.e. 2qter - end of the long arm
        of chromosome 2)
 tri	Trisomy
 trp 	Triplication of a portion of a chromosome 

The software CyDAS seems to be able to work with this kind of karyotype data and there is also a discussion about shortcomings of the nomenclature. (Some of the issues might have been fixed in the mean times.)

Sources:

• International Standing Committee on Human Cytogenetic Nomenclature (2005) publication
• Brothman et al. (2009) Nomenclature Evolution: Changes in the ISCN from the 2005 to the 2009 Edition
• Franke (1981) High-resolution ideograms of trypsin-Giemsa banded human chromosomes. and Franke (1994)
• WSHL
• Schreck et al (2001) ISCN standard idiograms
• Ideograms at WashU

This is the list of genomic regions that was analysed as the 1% of the human genome in the ENCODE pilot phase. (The main phase of ENCODE is looking at the entire human genome.) The coordinates are for assembly NCBI36 (hg18).
See also the entry about ENCODE and the UCSC pages.

A quick reminder of the specifications to connect to the public Ensembl mySQL databases:

user = "anonymous"

pass = ""

mysql commandline for connection:

Code: