Homework

1
Homework 1

• Problem 1 Find CpG island in human chromosome
  21
• The formal definition of CpG island is
• It is longer than 200 symbols
• The CG probability is 0.61/16.
• GC content(GC out of ATGC) 50
• ? Moving average problem.

2

• Also, the CpG island should be maximal, in the
  sense that if we extend one nucleotide to the
  right or left, it is no longer a CpG island.
• Two methods
• Moving average.. Create a window and slide to the
  left or right
• Divide the whole set into ,say, 10 units of nts.
  Then, calculate GC content, CpG probability, and
  sum up. Also, the consider the bordering nts.

3

• You use the hs_chr21.fa on the lecture page.
• Your output should contain the CpG location,
  including the contig number, starting and ending
  location in that contig.

4

• Sample from chromosome 21 of human
• gi27501117refNT_011512.7Hs21_11669 Homo
  sapiens chromosome 21 genomic contig
• CATGTTTCCACTTACAGATCCTTCAAAAAGAGTGTTTCAAAACTGCTCTA
  TGAAAAGGAATGTTCAACTC
• TGTGAGTTAAATAAAAGCATCAAAAAAAAGTTTCTGAGAATGCTTCTGTC
  TAGTTTTTATGTGAAGATAT
• TTCCATTTTCTCTATAAGCCTCAAAGCTGTCCAAATGTCCACTTGCAGAT
  ACTACAAAAAGAGTGTTTCA
• AAAGTGCTCAATGAAAAGGAATGTTCAGCTCTGTGAGTTAAATGCAAACA
  TCACAAATAAGTTTCTGAGA
• ....
• gi27486021refNT_011515.9Hs21_11672 Homo
  sapiens chromosome 21 genomic contig
• AAGCTTCTCAATTTCAGAAATCTTCGGCAGCTTGGGGACATTCAAGGTCA
  CCCTGGGCTCCCAAAGTCAC
• ACAATTCCATTGGCCACAGCCAAGTTTCCACGTCAGGAGGCTGTGG
  GCTGGGGGTGGCAGCACTGGGTCC
• TGG
• NT_0011512.7 is the contig number.

5
Problem 2

• Your are given prosite.dat, containing pattern
  information, and swprot.fas, the swissprot
  database in plain file format. You should
  construct a perl program which outputs the
  proteins containing the pattern in the command
  line input. For example, if your program name is
  parse.pl and you want to find all proteins in
  swissprot database containing the patter PS00592
• perl w parse.pl PS00592

6

• Should print out the corresponding proteins.
• The PS00592 entry in the prosite.dat
• ID GLYCOSYL_HYDROL_F9_1 PATTERN.
• AC PS00592
• DT DEC-1991 (CREATED) DEC-1992 (DATA UPDATE)
  JUL-1998 (INFO UPDATE).
• DE Glycosyl hydrolases family 9 active sites
  signature 1.
• PA STV-x-LIVMFY-STV-x(2)-G-x-NKR-x(4)-P
  LIVM-H-x-R.
• NR /RELEASE42.0,135850
• NR /TOTAL18(18) /POSITIVE16(16)
  /UNKNOWN0(0) /FALSE_POS2(2)
• NR /FALSE_NEG2 /PARTIAL1
• CC /TAXO-RANGE??EP? /MAX-REPEAT1
• CC /SITE11,active_site
• DR P05522, GUN1_PERAE, T P23666, GUN2_PERAE,
  T P28622, GUN4_BACS5, T
• DR P26221, GUN4_THEFU, T P22699, GUN6_DICDI,
  T P22534, GUNA_CALSA, T
• DR P23665, GUNA_FIBSU, T P10476, GUNA_PSEFL,
  T P26225, GUNB_CELFI, T
• DR P23658, GUNC_BUTFI, T P04954, GUND_CLOTM,
  T P26224, GUNF_CLOTM, T
• DR P37700, GUNG_CLOCE, T Q02934, GUNI_CLOTM,
  T P23659, GUNZ_CLOSR, T
• DR P22503, GUN_PHAVU , T
• DR P38534, GUNX_PRUPE, P

7

• AC accession number such as PS00592.
• Detailed information can be found from the
  prosite database.

8

• The pattern is
• PA STV-x-LIVMFY-STV-x(2)-G-x-NKR-x(4)-P
  LIVM-H-x-R.
• One of sample protein is
• GUN4_BACS5 (P28622) Endoglucanase 4 precursor
  (EC 3.2.1.4) (Endo-1,4-b
• MTRRWSFLVQCFTFKKKEGVRSRYMSDYNYVEVLQKSILFYEAQRSGKLP
  ESNRLNWRGD
• SGLEDGKDVGHDLTGGWYDAGDHVKFGLPMAYSAAVLAWTVYEYREAYEE
  AELLDDMLDQ
• IKWATDYFLKAHTGPNEFWAQVGDGNADHGWWGPAEVMPMNRPAFKIDEH
  CPGTEVAAQT
• AAALAAGSIIFKETDAPYAAKLLTHAKQLYAFADQYRGEYTDCVTNAQPF
  YNSWSGYIDE
• LIWGGIWLYLATNDQTYLNKALKAVEEWPKDWDYTFTMSWDNTFFLSQIL
  LARITKEKRF
• IESTERNLDYWSTGFVQNGKVERITYTPGGLAWLDQWGSLRYTANAAFLA
  FVYADWVSDQ
• EKKNRYQTFAIRQTHYMLGDNPQNRSYVVGFGKNPPMHPHHRTAHGSWSN
  QLTTPSSHRH
• TLYGPLVGGPNRQDQYTDDISDYVSNEVATDYNAAFTGNGAAVWSGQSKL
  PNFPPKEKVE
• DEFFVEAAVMSNDTTSTQIKAILYNRSGWPARSSQSLSFRYYVNLSEIFA
  KGFTDKDIQV
• TAVYNEGASLSPLTVYDASSHIYFTEIDFTGVAIFPGGESLHKKEIQFRL
  SAPNGANIWD
• ASNDYSFQGLTSNMQKTARIPVFDQGDLVFGTLPNK

9
(No Transcript)
10
Problem 2) List
11
PERL continued

• Object
• A class is a combination of variables and
  functions designed to emulate an object.

12
(No Transcript)