A Program For Automatic Pedigree Construction With pst
Contents
1. 1935 12 01 70 II 1 1965 08 06 2 1968 05 05 1 1990 12 01 2005 12 08 11 2 1995 03 24 2005 12 08 Figure 2 Example of the Typeset Pedigree in Russian Data File from Listing 7 16 FO AO 1 2 1 2 B3 C1 C2 C3 C4 Name Adam Sam John Jane Jack Mike Pam Jane John Ann 01 01 14 05 DoB male unknown male 1950 male 1950 female 1951 male 1975 male 1975 female 1973 female 1998 male 1998 Georgelmale 1998 female 2003 05 11 DoD unknown 03 unknown 15 06 06 01 12 12 12 02 04 04 04 04 Mother Father 1 0312005 04 12 2 2 2 1 1 1 181 181 181 181 Listing 8 Example of Data File with Twins Adam IV 4 A1 qzygotic AO qzygotic B2 monozygotic B1 monozygotic IC2 C3 IC1 C3 IC1 C2 Figure 3 Example of a Pedigree with Twins Data File from Listing 8 17 10 B1 B2 B3 Name Ann John Sex DoB DoD femal2. T0 ZT G00 sak Tal 1G0 Z0 OL6T uyor d 3au umoo 0 qoq xeg eueN Listing 7 Examples of Data Files English and Russian 14 Joseph Smith Joan Smith Jim Brown Lisa Brown 1 2 1 4 Bill Smith Jenny Smith Rebecca Smith Alexander Smith Z 11 3 11 4 I 1 Joseph Smith born unknown age at death unknown I 2 Joan Smith born 1902 07 01 age at death 73 I 3 Jim Brown born 1905 11 01 I 4 Lisa Brown born 1910 03 03 Ann Gold born 1941 09 02 Aunt of the proband II 2 Bill Smith born 1938 04 03 II 3 Mary Smith born 1940 02 05 Jenny Smith born 1969 12 03 Cousin of the proband IIL 2 John Smith born 1970 02 05 Evaluated 2005 12 01 IILE3 Rebecca Smith born 1972 12 25 III 4 Alexander Smith born 1975 11 12 Figure 1 Example of the Typeset Pedigree in English Data File from Listing 7 15 1 2 III 1 III 2 I 1 1 2
3. descarmA in descarmA 0 8 Distances between nodes in cm xdist 2 ydist 2 Listing 5 Configuration File Choosing Lengths The scaling works as follows If both height and width of the pedigree are smaller than the limits no scaling is done In the other case the chart is scaled while preserving the aspect ratio by changing the value of unit see 3 to fit into the limits The variable rotate sets the orientation of the chart If it is no the pedigree is never rotated while if it yes it is always rotated ninety degrees counterclockwise If this variable is set to maybe the default the program compares the scaling for the non rotated and rotated pedigrees and chooses the orientation for which the scaling is closer to one 4 Running the Program 41 Program Invocation And Options The program pedigree is a command line program It reads the data from a text file input_file and produces an output file with macros The format of the input file is described in Section 4 2 The program invocation is pedigree c configuration_file d o output_file s start input file the square brackets show optional arguments All arguments but input file are optional They are described below option c selects a configuration file format of the configuration file is described in Section 3 1 If this option is absent the program uses its own default parameters or system
4. esonegy ex 80 51 3005 oom d T0 ZI O66T X x Tal THI 80 80 8961 182 10 200 gouegy Ta IN er THI 190 80 8961 KAN d 2000 dex TKO pueqoid eua Fo ursno 20 21 6961 yatug Auuer T pueqoad uq jo uny 8 ta 1 20 60 1561 uuy TV 121 11 9261 ZS Tal THI 92 1 061 TS Tewzou 0 0 0T6T ereueg umoxg TO TT SO6T unmoag 249 umouxun umouyun udeso T45 T ZT S2Z6T 10 40 2061 weor TW Ta 10 0 v0 8 6T TITS ld Teuzou 42 ZWD 80 20 0761 xew
5. 1 1 G1 G2 Hi K1 M1 P1 R1 X1 Name John Sam Ann Paul Scott Simon Sarah Lola Jim Jane Simon Pam James Figure 8 Pedigree from Listing 13 Sex DoB male 1915 male 1935 female 1937 male 1952 male 1912 male 1934 female 1936 female 1960 11962 female 1917 11935 female 1964 11988 male male male 06 12 02140 10 02 09 12 04 03 11 IV 2 Father Mother Proband 15 04140 03 B1 011 17 F1 19 F1 13162 05 G1 02 10 02 07 13 04 05 1 12 K1 24 yes John Listing 14 A Pedigree with Unavoidable Self Intersections I euer Figure 9 Pedigree from Listing 14 25 Name Father Mother AO Jane female 11908 12 12 John male AO 1936 12 15 B2 Ann female AO 1934 04 17 Samantha female AO 1932 12 03 B4 1 1928 01 05 C1 Mary female B2 yes 1955 08 26 C2 Paul male B3 1964 05 07 Jane female B4 1950 11 03 1 male 1 C1 11975 07 01 02 Laura female C2 c3 1974 09 05 Listing 15 Pedigree with Consanguinic Unions 4 7 Consanguinic Unions Consanguinic unions present a technical problem for the program see the dis cussion in Section 8 Therefore the support of consanguinicity is exp
6. The field names are described in Section 4 2 Note that AgeAtDeath is a special field it is the age at death or empty calculated as the difference between the death date and the birth date 3 5 Language and Encoding The next group of variables describes the language and encoding of the data file input and the IATEX output They are shown in Listing 3 The variable language at present can have one of two values english the default or russian If the value is russian the output document preamble includes the line usepackage russian babel The variable encoding sets the encoding of the file if the language is not English By default it is cp1251 if the language is Russian Set it to koi8 r to choose KOIS encoding It is worth to note that the data file and the output IXTEX file are assumed to have the same language and encoding If 1anguage is not english the program recognizes both English and native names of the fields in the data file see Section 4 2 3 6 Fonts There are two kinds of text on the chart the text above a node and the text below a node The fonts for them are set by the variables belowtextfont by default small and abovetextfont by default Nscriptsize Any font declaration like sffamily or itshape is allowed here See Listing 4 for an example of usage 4The TEX package 2 also allows to place text at both sides of the node but the program pedigree currently does not use this f
7. Sex DoB DoD Mother Father Proband Condition Comment Type Twins SortOrder Field Values male female unknown yes no normal obligatory asymptomatic affected infertile sab monozygotic qzygotic Special Names abortion childless Sort obligat asymp affect monzygot qzygot 2 Table 1 Keywords in Different Languages 28 Part II Algorithm Description 5 Introduction This part is intended for advanced users and is not neccessary for runnuing the program problem of nicely typesetting graphs is one of the classical problems in the Computer Science 4 One of the earliest algorithms here is the classical algorithm for layered rooted trees by Reingold and Tilford 4 8 3 1 This algorithm was implemented by PSTricks 3 However many pedigrees are not trees 2 If we consider a subset of pedigrees where inbreeding is absent the pedigrees become trees However even in this case the the tree is not necessary layered as
8. on the diagram While this is relatively easy for small pedigrees this task becomes increasingly time consuming for larger ones There may be several approaches to automate it For example one may have data about the patients and their families in spreadsheet or database Then it would be useful to generate pedigrees from such data This is the aim of the program pedigree described in this manual Spreadsheets and databases can export the data as separated values files csv files for Comma Separated Values Our program reads these files and outputs LaTeX code with pst pdgr macros We tried to make this code read able so a user might tweak it if necessary Of course manually produced IATEX code is more versatile than the auto matically generated one There are certain limitations for the program 1 only persons having common genes with the proband or the starting person are included in the pedigree 2 no adopted children sperm donors or surrogate mothers are shown on the pedigree 3 only one disease is shown on the chart 4 the support for consanguinic unions and inbreeding is rather experimental see Section 4 7 Subsequent versions of the program may ease some of these limitations 2 Installation 21 System Requirements The program requires Perl version 5 or newer it was tested with Perl v5 8 8 but should work with any Perl 5 The IATEX macros require pst pdgr version 0 3 July 2007 or newer 2 2 Unix Linux Install
9. 1944 male 1941 male 1963 male 1961 female 1965 female 1965 11989 female 1991 male Listing 11 A Data File with a Sorting Problem Sex DoB male 1915 female 1940 female 1942 male 1944 male 1941 male 1963 male 1961 female 1965 female 1965 11989 1 1991 male Father Mother Proband 15 02140 07 AO 01140 011 01184 26 12 11 21 C1 02 C1 Father Mother Proband SortOrder 15 02140 07 AO 01140 011 01184 26 12 11 21 C1 02 C1 21 B2 B1 B3 4 104 B2 B1 B3 C4 IC4 Listing 12 First Solution to the Problem in Listing 11 Figure 6 Pedigree from Listing 12 22 10 B1 B2 B3 B4 1 C2 C3 C4 Di D2 Name John Joan Jane Bill Peter Jack Sam Ann Lucy Mark Dina John Figure 7 Pedigree from Listing 12 Sex DoB male 1915 female 1940 female 1942 male 1944 male 1941 male 1963 male 1961 female 1965 female 1965 11989 1 1991 male Listing 13 Second Solution to the Problem in Listing 11 06 03 07 01140 05 12 08 12 11 12 Father Mother Proband SortOrder 15 02140 07 01 01184 26 12 12 06 021 1 111 21 C1 23 B2 B1 B3 104 104 IV 1 IV 2 10 1 2
10. 4 ADBOELUODS s i 4 5 Childlessness and Infertility 4 6 Ordering Siblings and Marriage Partners 4 7 Consanguiic Unions 2 222 4 8 Language Dependent Keywords II Algorithm Description 5 6 7 8 9 Introduction Main Algorithm Algorithm for Sorting Siblings and Marriage Partners Modifications for Consangunic Unions Conclusion 10 Acknowledgements A A gt gt A 0000 10 orc 29 29 29 30 31 31 32 List 1 gt List List CONDO gt C2 h2 of Figures Example of the Typeset Pedigree in English Data File from List nrc ate ote 15 Example of the Typeset Pedigree in Russian Data File from Listing Tires mud i Ba a Ao othe Oe 16 Example of a Pedigree with Twins Data File from Listing 8 17 Example of a Pedigree with Abortions Data File from Listing 9 18 Example of a Pedigree with Childlessness Data File from List Ing ia occa en hE ee di Ag s dead ter Bans usate 20 Pedigree from Listing 12 22 Pedigree from Listing 12 23 Pedigree from Listing 13 24 Pedigree from Listing 14 25 Pedigree from Listing 15 0 27 Subpedigr
11. A Program For Automatic Pedigree Construction With pst pdgr User Manual and Algorithm Description Boris Veytsman borisvOlk net Leila Akhmadeeva March 2012 Abstract The set of macros in pst pdgr package allows to typeset complex pedi grees However a manual placement of pedigree symbols on a canvas is time consuming task This program produces TFX files from spreadsheets with the data on inheritance for a large class of pedigrees It has a simple interface and can be used for quite complex pedigrees 1 Z 2 11 3 11 4 Contents I User Manual 1 2 Introduction Installation 21 System Requirements 2 2 Unix Linux Installation icp 2c wee eva 2 3 Installation in Other Systems Configuration 3 1 Configuration Variables and Location of Configuration File 3 2 Configuration File amp 3 3 TPX Output Setup 3 3 10v OR OR ee fers 3 45 What to Print amp bh bE are pue ee 3 5 Language and Encoding 3 6 Bonis ee OA oe ee A 34 Lengths on orgie A A S Q 3 8 Scaling and Rotation st RI uw Running the Program 41 Program Invocation And Options 42 Data Sehr we BNR m Ee eS BR he es ate ha H 4 8 TWINS cided Rated oe Bint me eb Gd Be oad GAR MM Z
12. ate of death use the keyword unknown Note the subtle difference between the fields DoB and DoD an empty value for DoB is means unknown birth date while for DoD it means that there is no date of death at all Mother The Id of the mother of the person or empty Father The Id of the father of the person or empty Proband This field can be either yes for the probands or empty or no for other persons Note that if a pedigree has no probands or several probands the program does not know from which node to start the pedigree There fore in this case the option s must be used to explicitly set the Id of the starting chart node see Section 4 1 12 Condition This column can have the values normal obligatory asymptomatic or affected If it is empty the default value normal is assumed Comment A comment about the person Twins If the current person has twins they are listed in this column separated by spaces and or commas See Section 4 3 for more details Type This column is used in certain special cases For abortions it shows the type of the abortion Section 4 4 for childless people and marriages it shows the type of childnessness Section 4 5 and for twins it shows the type of twins Section 4 3 SortOrder This column is used when the algorithm for sorting siblings and unions gives a wrong result and a manual correction is needed See Sec tion 4 6 for the explanation and examples Examples of data fil
13. ation If your system has a working make program which is the usual case for Unix like environments the supplied Makefile installs the executable pedigree in usr local bin the libraries in usr local lib site_perl and the manual pages in usr local man This is done by the usual command make install Optionally you can install files in the doc and examples subdirectories in the proper places in your system 2 3 Installation in Other Systems If your system does not have make you need to manually perform the following 1 Install the executable pedigree pl to the place your system can find it 2 Install the libraries Pedigree pm directory Pedigree and all files in it to the Perl search path The latter is listed in the array INC which can be checked by the command perl V or its equivalent 3 Configuration 31 Configuration Variables and Location of Configura tion File The program defaults are sufficient for most cases However if you want to draw pedigrees in a language other than English or to tweak the layout of the pedigrees you need to change the program configuration The behavior of the program pedigree is determined by configuration vari ables There are several sources of configuration variables They are in the order of increasing priority 1 Program defaults 2 The system configuration file etc pedigree cfg On TpEXLive the sys tem coniguration files are TEXMFHOME texmf config pedigree pedigree cfg an
14. can be seen from Figure 1 Therefore a new approach generalizing Reingold Tilford algorithm is necessary This approach is based on the analysis of the structure of pedigrees and is sketched in the remainder of this manual 6 Main Algorithm pedigree consists of nodes vertices connected by lines edges If there is no inbreeding the graph is acyclic There are two kinds of nodes in the graph person nodes squares and circles on Figures 1 and 2 and marriage nodes which are nameless on the figures We will use the notation male spouse female spouse for such nodes so the marriage nodes on Figure 1 are 1 1 1 2 3 4 and 2 3 node has a precedessor and children A marriage node does not have a precedessor but has male spouse and female spouse it is customary to put male spouses to the left and female spouses to the right on pedigrees Any node has a downward tree of its children grandchildren etc The downward tree may be empty Any node in an acyclic graph can be a root However in layered trees there is a special root the one that has no precedessor Similarly we will call a local root a node that has no predecessor All marriage nodes are local roots Some person nodes can be local roots as well Let us first discuss the case where cobnsanguinic marriages are absent In this case a pedigree is a tree The proposed algorithm is recursive and starts from a local root Strictly speaking it can start from any local ro
15. column headers in the first line All fields but Id are optional If the value is empty for all rows the corre sponding column can be dropped If applicable the default values for this field will be substituted by the program On the other hand the data file can include any additional columns as long as their names do not clash with the names listed below and the special name AgeAtDeath These additional columns can be included in the chart or legend as described in Section 3 4 Here is the list of columns and explanation of their meaning Id Each line including the special lines described below must have a unique Id The Id may contain only Latin letters and numbers and start with a letter Name The name of the person described in the current row There are also special names when the current row describes abortions or infertility They are described below The names should not contain special symbols like 5 etc Sex The gender of a person This column may have one of two values male or female The empty value corresponds to a person with unknown gender DoB The date of birth for the current person The format is YYYY MM DD If the date of birth is not known the field may be empty or the keyword unknown may be used DoD The date of death for current person The format is the same as for DoB YYYY MM DD If this field is empty the corresponding person is alive For deceased persons with an unknown d
16. cument when fulldoc 0 The default is 1 The variable documentheader is used when fulldoc is 1 It determines the document class of the resulting IXTEX file The default is article class set by Ndocumentclass article By default the preamble of the file created when fulldoc is 1 con tains only the line usepackage pst pdgr and if the language chosen is not English see Section 3 5 the calls of babel and inputenc packages The variable addtopreamble if set may contain any other code you might wish to add to the preamble The variable printlegend determines whether to add legend to the pedi gree The default value is 1 and the legend is printed 3 4 What to Print The next groups of configuration variables sets the information to be printed in the legend and on the pedigree It consists of two arrays array fieldsforlegend is the list of fields see Section 4 2 which are included in the legend and array fieldsforchart is the list of fields to print near each node in the pedigree Listing 2 Setting fieldsforchart to empty array fieldsforchart Fields to include in the legend Delete Name for privacy protection fieldsforlegend qw Name DoB DoD Comment Fields to put at the node Delete Name for privacy protection fieldsforchart Listing 2 Configuration File Choosing Fields to Print prevents putting additional information on the pedigrees
17. d TEXMFLOCAL pedigree pedigree cfg 3 User configuration file HOME pedigreerc 4 The file specified by the option see Section 4 1 If a file mentioned in this list does not exists the program silently continues Note that even if a configuration file with higher priority exists the program reads the files with lower priority first The former overrides the latter but not precludes it from reading In other words if etc pedigree cfg defines variables foo and bar and HOME pedigreerc defines bar and baz the program takes foo from the first file and bar and baz from the second one 1On Unix like systems where etc exists 2On Unix like systems where HOME exists 3Unless d option is selected see Section 4 1 3 2 Configuration File Format All configuration files mentioned in Section 3 1 have the same format They are actually snippets of Perl code executed by the program pedigree This means by the way that all precautions usually taken with respect to programs and scripts are relevant for configuration files as well In particular it is a bad idea to have world writable system wide configuration file etc pedigree cfg The code in configuration files is very simple and one does not need to know Perl to edit configuration files There are several simple rules which are enough to understand these files 1 All text after to the end of the line is a comments In particular the lines starting with are c
18. e 1970 06 15 abortion female 1990 03 01 abortion male male 2000 10 10 2002 12 01 Mother Proband Condition Type affected laffected sab AO lyes laffected Listing 9 Example of Data File with Abortions I 1 Ann born 1970 06 15 1 abortion born 1990 03 01 2 abortion born 2000 10 10 II 3 John born 2002 12 01 Z H Figure 4 Example of a Pedigree with Abortions Data File from Listing 9 Name Sex Mother Father Proband Type Comment AO John male James Imale AO Bic childless male B1 linfertile anospermia B2 Ann female AO lyes B2c childless B2 l Listing 10 Example of Data File with Childlessness Father The Id of the childless male If both Mother and Father columns are not empty the entry describes the union between the Father and Mother Of only Mother or Father is not empty the entry describes the state of the corresponding person Type This column might be either empty or have a keyword infertile In the latter case the childlessness of the person or union is caused by a proven infertility Comment The vaule of this column is shown under the childlessness symbol on the chart Put there a short description of the cause of childlessness like anospermia or vasectomy An example of a pedigree with childlessness is shown on Listing 10 and Figure 5 4 6 Ordering Siblings and Marr
19. eature Language language russian language english Override the encoding encoding koi8 r Listing 3 Configuration File Choosing Language and Encoding Fonts for the chart belowtextfont small abovetextfont scriptsize Listing 4 Configuration File Choosing Fonts 3 7 Lengths The next group of variables Listing 5 sets the distances between the key ele ments of the chart All lengths are in centimeters actually in units are defined in PSTricks 3 The variable descarmA sets the length of the first segment of the descent line from the parent node to the sibs line as measured from the center of the parent see 2 for more details By default it is 0 8 The variables xdist and ydist set the distances between the nodes along horizontal and vertical axes correspondingly The default for both is 2 3 8 Scaling and Rotation Complex pedigrees might be too large to fit on a page In this case a scaling and or rotation might be necessary to print the chart Of course changing the lengths described in Section 3 7 might also help but the scaling described here also changed the size of the pedigree symbols There are three variables controlling the scaling and rotation of pedigrees maxW maxH and rotate see Listing 6 The variables maxW and maxH are the maximal width and height of the chart in centimeters Setting any of them to zero disables scaling
20. ees and Downward 30 of Tables Keywords in Different Languages 28 of Listings Configuration File Setting TEX Output 7 Configuration File Choosing Fields to Print 8 Configuration File Choosing Language and Encoding 9 Configuration File Choosing Fonts 9 Configuration File Choosing Lengths 10 Configuration File Choosing Scaling and Rotation 11 Examples of Data Files English and Russian 14 Example of Data File with 17 Example of Data File with Abortions 18 Example of Data File with Childlessness 19 A Data File with a Sorting Problem 21 First Solution to the Problem in Listing 11 21 Second Solution to the Problem in Listing 11 23 A Pedigree with Unavoidable Self Intersections 24 A Pedigree with Consanguinic Unions 26 Part I User Manual 1 Introduction Medical pedigree is a very important tool for clinicians genetic researchers and educators As stated in 1 construction of an accurate family pedigree is a fundamental component of a clinical genetic evaluation and of human genetic research The package pst pdgr 2 provides a set of PS Tricks macros see 3 to typeset pedigrees In the framework of pst pdgr the user manually chooses coordinates for each pedigree node
21. enet 56 3 745 752 1995 Boris Veytsman and Leila Akhmadeeva Creating Med ical Pedigrees with PSTricks and July 2007 http ctan tug org tex archive graphics pstricks contrib pedigree pst pdgr 3 Timothy Van Zandt PSTricks PostScript Macros for Generic TEX July 2007 http ctan tug org tex archive graphics pstricks base doc 4 Giuseppe Di Battista Peter Eades Roberto Tamassia and Ioannis Tollis Graph Drawing Algortihms for the Visualization of Graphs An Alan R Apt Book Prentice Hall New Jersey 1999 32
22. erimental for this release There is a number of limitations for consanguinic unions in the data file at present First the consanguinic unions should not in the direct lineage of the proband or the person from which the pedigree starts In many cases this limi tation can eliminated by using 8 option see Section 4 1 to choose a different starting point for the pedigree Second the children of consanguinic unions might appear not centerd on the charts An example of a pedigree with consan guinic marriages is shown on Listing 15 and the corresponding chart is shown on Figure 10 The drawbacks of the program are evident from the positions of Laura nad Jack on these charts 4 8 Language Dependent Keywords At present the program pedigree can work with English and Russian languages As discussed in Section 3 5 the language options chooses both the languages of input and output files It is easy to add new languages to the scheme by expanding the library Pedigree Language pm in the distribution English language is the default Moreover if the Russian option is chosen English keywords are still recognized in the input file The English and Russian keywords are listed in Table 1 Note that some keywords have variants they are listed in the table as well 26 11 2 Z IIL3 IV 1 IV 2 Figure 10 Pedigree from Listing 15 27 John English keyword English variants Russian keywords Field Names Id Name
23. es in English and Russian are shown in Listing 7 the Russian keywords are discussed in Section 4 8 4 3 Twins The column Twins see Section 4 3 lists all Ids of all twins of the given person The column Type can be used to show the type of the twins The empty value means polyzygotic twins monozygotic means monozygotic twins and qzygotic is used in the case when the type of twins is under doubt An example of a data file with twins is shown on Listing 8 and the corresponding pedigree on Figure 3 4 4 Abortions Aborted pregnancies are described by a special entry in the data file The field Name has the value abortion the symbol is used to show that this is a special value The columns Sex DoB Mother Father and Condition have the usual meaning The special column Type is either empty or be equal to sab for self abortions 4 5 Childlessness and Infertility Childlessness is can be a property of a person or a union between two persons Therefore in this implementation we use a special row rather than a column to report it As other rows this one has a unique Id The Name column should have a special entry childless Like abortion Section 4 4 this special name starts with to distinguish it from real names There are four other columns that have meaning for this row Mother The Id of the childless female 13 80 51 3005 d 172 20 9661
24. iage Partners The generations in pedigrees are ordered in vertical direction from up do down How should we order the people on the same generation i e siblings and mar riage partners Usually two rules are used 1 The siblings are ordered from the oldest on the left to the youngest to the right 2 In marriage or other union the male is to the left and the female is to the right However the combination of these rules might lead to the situation when mar riage lines intersect the parental lines Therefore the rule 1 is usually implicitly modified la The are ordered from the oldest on the left to the youngest to the right However if a sibling s marriage is shown on a pedigree this sibling is always the rightmost male or the leftmost female 19 n2 anospermia Figure 5 Example of a Pedigree with Childlessness Data File from Listing 10 The program follows these rules It is enough to draw pedigrees in most cases In particular they always produce correct pedigrees if there is only one mar riage shown However in complex cases these rules fail as shown on Listing 11 and Figure 6 It is possible to extend the rules above to account for these cases however we chose another solution to provide a facility for the manual intervention in the sorting and ordering algorithm For this purpose a special column SortOrder is used It can have positive numbers greater than 1 or negative nu
25. le to be the rightmost and the female to be the leftmost To do so we assign to each node the special quantity SortOrder Initially all nodes have SortOrder equal to zero unless specifically set by the user in the input file see Section 4 6 Then we use the following rules 1 When creating the the marriage node 30 If both spouses have equal SortOrder field the male goes to the left the female goes to the right b Otherwise the spouse with greater SortOrder goes to the left c If SortOrder of a spouse is 0 we set it to 1 the spouse on the left or 1 the spouse on the right 2 When sorting siblings a The sibling with smaller SortOrder goes to the left b If both siblings have the same SortOrder the oldest one goes to the left 8 Modifications for Consangunic Unions Consanguinic unions present a problem for the described algorithm because pedigrees with them are no longer trees see Figure 10 In this release of the program we use the following hack The direct lineage of the proband or more generally the starting node may have both mothers and fathers in the pedigree because they share genes from the starting node If any other person has both mother and father in the chart his or her parents both shared their genes with the starting node Therefore they formed a consanguinic union In this case the children of this node appear in two subtrees their mother s and their father s We delete
26. mbers smaller than 1 If the value of this column is positive the corresponding person is moved to the left when sorting siblings and to the right when sorting marriage partners If it is negative the opposite sorting rule is ap plied see Section 7 for more detailed discussion Note that sibling sorting and marriage partners sorting must work in opposite directions otherwise marriage lines intersect paternal lines Let us return to the pedigree on Listing 11 To improve Figure 6 we can either move Peter to the right or Lucy to the left The first solution is shown on Listing 12 and Figure 7 The second is shown on Listing 13 and Figure 8 Of course sometimes a pedigree cannot be drawn without self intersections with any sorting of siblings An example of such pedigree is shown on Listing 14 and Figure 9 Obviously no amount of shuffling the siblngs can help in his case If the program cannot avoid self intersection of marriage lines and parental lines despite automatics sorting and manual intervention as the last resort it creates a multi segment marriage line as shown on Figures 6 and 9 20 10 1 2 B3 B4 C1 C2 C3 C4 Di D2 Id 10 1 2 B3 B4 C1 C2 C3 C4 Di D2 Name John Joan Jane Bill Peter Jack Sam Ann Lucy Mark Dina Name John Joan Jane Bill Peter Jack Sam Ann Lucy Mark Dina Sex DoB male 1915 female 1940 female 1942 male
27. omment lines 2 Perl commands must end by semicolon 3 The commands like xdist 1 5 or fieldsforprint qw Name assign values to the variables 4 Variables starting with are scalars and take numerical or string values Variables starting with are arrays and take list of values 5 A backslash in single quotes stands for itself A backslash in double quotes or inside lt lt END END construction must be doubled Compare the com mands fo0 documentclass bar documentclass 6 The last command in the file must be 1 A number of commented configuration files can be found in the examples subdirectory of the distribution In the remaining parts of this section we describe the configuration variables in detail 3 3 Output Setup A number of variables determine what kind of file is produced An example of their usage is shown on Listing 1 Do we want to have full LaTeX file or just a fragment fulldoc 1 What kind of document do we want documentheader documentclass farticle Define additional packages here addtopreamble lt lt END usepackage pst pdgr END Do we want to print a legend printlegend 1 Listing 1 Configuration File Setting Output The variable fulldoc determines whether the program produces a full IATEX file with header and preamble when fulldoc 1 or just a snippet to be included in a larger do
28. ot but medical pedigrees have a special person proband the person who was the first to be examined by genetic spe cialists the proband is shown by an arrow drawn near the node on Figures 1 and 2 Therefore it makes sense to start from the local root which has proband in its downward tree If this local root is person node the pedigree is the layered tree and Reingold Tilford algorithm is sufficient Therefore we should consider only the 29 Left subpedigree Right subpedigree ss Ps Local E m Downward tree Figure 11 Subpedigrees and Downward Tree case when the local root is a marriage node In this case we can typeset the downward tree using Reingold Tilford algorithm The spouses do not belong to this tree However each of them belongs to each own subpedigree We will call them left subpedigree and right subpedigree We recursively apply our algorithm to typeset left and right subpedigrees Then we move the left subpedigree to the right and right subpedigree to the left as far as we can without intersection between them and the downward tree This process is shown on Figure 11 Obviously this algorithm converges and leads to typesetting the pedigree without intersections between the subtrees and subpedigrees 7 Algorithm for Sorting Siblings and Marriage Partners When we create a marriage node we want to put the male to the left and the female to the right When we then sort siblings we want this ma
29. s special The invocation pedigree v outputs the version and license information 4 2 Data File The input for the program is a separated values file Usually such files are called CSV for comma separated values However this program uses the vertical bar pipe as a separator Each line of this file is a record The lines are separated by pipes into fields Most SQL programs produce such files by default Spreadsheet programs will make them if you choose Save As option and select as the field separator and empty text delimiter We sometimes will call the records rows and the fields columns to use the familiar spreadsheet metaphor Normally each row corresponds to a person in a pedigree We will call this person the current person when describing the fields 5The proband is the first person among the relatives who came to a geneticist he or she is the primary patient 11 The width of the fields may not be the same in all rows in other words the pipes may be disaligned We make them aligned in the examples included in this manual just to make the text more readable The first line of the data file contains the names of the fields column headers The fields in the subsequent lines must match the order of the headers An empty field must be still included as Otherwise the order of columns is arbitrary as long as it is the same for all rows i e matches the order of
30. them from one of the subtrees the one with lower generation number connect their parents with a double line consanguinic union and put the descent line from the middle of the union to them There are two problems with this hack see Section 4 7 the children of consanguinic unions are not centered on the diagaram and the hack fails if the starting node itself is a descendant of a consanguinic union Probably the next releases will employ better algorithms for consanguinic unions 9 Conclusion The algorithm seems to be efficient and producing nicely typeset pedigrees Since the input file format is simple it may be used by the people without special skills in IATEX On the other hand the files produces are easy to understand and edit manually if the need arises 31 10 Acknowledgements The authors are grateful Herbert for help with PSTricks code The support of User Group is gratefully acknowledged One of the authors LA was supported by Russian Foundation for Fundamental Research travel grant 06 04 58811 Russian Federation President Council for Grants Supporting Young Scientists and Flagship Science Schools grant MD 4245 2006 7 References 1 Robin L Bennett Kathryn A Steinhaus Stefanie B Uhrich Corrine K O Sullivan Robert Resta Debra Lochner Doyle Dorene 5 Markei Vic toria Vincent and Jan Hamanishi Recommendations for standardized hu man pedigree nomenclature Am J Hum G
31. wide or user s defaults as explained in Section 3 1 The option d selects debugging mode In this mode a lot of debugging messages are dumped to stderr The parameter o provides the name of the output file Both input file and output file can be which means stdin for the input and stdout for the output If the parameter o is absent the program tries to guess the name 10 Maximal width and height the pedigree in Set this to 0 to switch off scaling maxW 15 maxH 19 Whether to rotate the page The values are yes no and maybe If maybe is chosen the pedigree is rotated this provides better scaling rotate maybe Listing 6 Configuration File Choosing Scaling and Rotation of the output file from the name of the input file If the input file is foo csv the output file will be foo tex On the other hand if the input file is stdin the output file is stdout Usually pedigrees are built starting from the proband Only the people that share genes with the proband are shown on the pedigree However in some cases for example when there is no proband or where there are several probands it is neccessary to override this default and tell the program from which person to start This is done using the option s If it is present it must be followed by the Id of a person in the data file see Section 4 2 for the discussion of Id The option i
Download Pdf Manuals
Related Search