;http://www.gene.ucl.ac.uk/nomenclature/data/gdlw_columndef.html

(setq *families* (make-hash-table :test 'equal))

(defun each-hugo-synonyms (function)
  (and (config :hugo)
       (with-open-file (f (config :hugo))
	 (let ((headers (mapcar 
			 (lambda(h) (intern (substitute #\- #\space 
							(regex-replace-all "[()]" (string-upcase h) "")) 'keyword))
			 (split-at-char (read-line f) #\tab))))
	   (let ((to-reap (list (list (position :previous-symbols headers) :multiple)
				(list (position :approved-name headers) :single)
				(list (position :approved-symbol headers) :single)
				(list (position :previous-symbols headers) :multiple)
				(list (position :previous-names headers) :multiple)
				(list (position :aliases headers) :multiple)))
		 (entrez-pos1 (position :entrez-gene-id headers))
		 (entrez-pos2 (position :entrez-gene-id-mapped-data headers))
		 (gfn (position :gene-family-name headers)))
	     (flet ((split-multiple (s)
		      (and s
			   (let ((split (remove nil (mapcar 'car (mapcar (lambda(s) (split-at-char s ", ")) (split-at-char s #\"))))))
			     (if (equal (car split) "")
				 (cdr split)
				 split))))
		    (not-empty (s)
		      (if (or (null s) (equal s "")) 
			  nil
			  s)))
	       (loop for line = (read-line f nil :eof)
		  until (eq line :eof)
		  for fields = (split-at-char line #\tab)
;		  for paired = (loop for h in headers for f in fields collect (list h f))
		  for entrez = (or (not-empty (nth entrez-pos1 fields))
				   (not-empty (nth entrez-pos2 fields)))
		  for names = 
		  (loop for (pos quant) in to-reap
		     for names = (if (eq quant :single) 
				     (list (nth pos fields))
				     (split-multiple (nth pos fields)))
		       append names)
		  for fn = (nth gfn fields)
		  when (and entrez names) do (funcall function entrez names)
		  when (and entrez (not-empty fn))  do (push entrez (gethash fn *families*)))
	     ))))))

#|
((:HGNC-ID "6871") (:APPROVED-SYMBOL "MAPK1")
 (:APPROVED-NAME "mitogen-activated protein kinase 1")
 (:STATUS "Approved") (:LOCUS-TYPE "undef")
 (:PREVIOUS-SYMBOLS "PRKM2, PRKM1") (:PREVIOUS-NAMES "")
 (:ALIASES "ERK, ERK2, p41mapk, p38, p38, MAPK2")
 (:CHROMOSOME "22q11.2") (:DATE-APPROVED "05/11/1993")
 (:DATE-MODIFIED "23/09/1999") (:DATE-NAME-CHANGED "")
 (:ACCESSION-NUMBERS "M84489") (:ENZYME-IDS "") (:ENTREZ-GENE-ID "")
 (:MGD-ID "") (:MISC-IDS "") (:PUBMED-IDS "") (:REFSEQ-IDS "")
 (:GENE-FAMILY-NAME "") (:GDB-ID-MAPPED-DATA "GDB:135677")
 (:ENTREZ-GENE-ID-MAPPED-DATA "5594") (:OMIM-ID-MAPPED-DATA "176948")
 (:REFSEQ-MAPPED-DATA "NM_002745") (:UNIPROT-ID-MAPPED-DATA "P28482"))
|#