Postgresql : Request optimization

[ikit]

The following query takes long time. Is it possible to optimize it ?

UPDATE wt_1 AS w SET is_exonic=True FROM refgene_exon_hg38 AS r WHERE w.chr=r.chr AND w.pos <@ r.exonrange

(venv) ➜  regovar git:(master) ✗ sql regovar "explain update wt_1 as w set is_exonic=True from refgene_exon_hg38 as r where w.chr=r.chr AND w.pos <@ r.exonrange" 
                                                        QUERY PLAN                                                        
--------------------------------------------------------------------------------------------------------------------------
 Update on wt_1 w  (cost=0.29..46206757.52 rows=23403374 width=1114)
   ->  Nested Loop  (cost=0.29..46206757.52 rows=23403374 width=1114)
         ->  Seq Scan on wt_1 w  (cost=0.00..57623.45 rows=681645 width=1108)
         ->  Index Scan using refgene_exon_hg38_exonange_idx on refgene_exon_hg38 r  (cost=0.29..66.33 rows=137 width=32)
               Index Cond: (w.pos <@ exonrange)
               Filter: (w.chr = chr)

refgene_exon_hg38 is the table with the list of all exons (658213 rows, with index on (chr, exonrange))

CREATE INDEX refgene_exon_hg38_chrom_exonange_idx
  ON refgene_exon_hg38
  USING btree (bin, chr, exonrange);
  
CREATE INDEX refgene_exon_hg38_exonange_idx
  ON refgene_exon_hg38
  USING gist (exonrange);

wt_1 is the working table created for an analysis (here the analysis with the id=1), with all variant s and their annotations