he generation of multiple sequence alignments and on the scanning of these alignments to identify polymorphisms. As mentioned, the sequence of the T. cruzi genome was obtained using a whole genome shotgun strategy, from a hybrid clone. Because of the sequence divergence between alleles of the CL Brener clone, assembly of this genome resulted in many cases in the separation of these alleles into separate contigs. This allowed us to align these sequences and identify sequence differences. However, because of the repetitive nature of the T. cruzi genome, we decided to focus this initial effort on mapping the genetic diversity in mostly single copy protein coding loci. These were defined as those sequences repre sented by no more than 2 coding sequences from the CL Brener genome in our sequence alignments.

Sequences used in this work include all the annotated coding sequences from the reference CL Brener genome, and the corresponding coding sequences from the Sylvio X10 genome, as well as other publicly available sequence data. After clustering sequences by similarity we obtained 7,639 multiple se quence alignments, 71. 3% of which had 2 reference coding sequences from the CL Brener Drug_discovery genome. Other alignments contain increasing numbers of reference coding sequences. These set of alignments contains sequences for most of the large gene families of T. cruzi, and were not considered further. Even after this stringent filte ring, there were still a number of alignments that contained only two reference sequences from the CL Brener genome, but that belonged to these large gene families mucins, mucin associated proteins, trans sialidase like proteins, etc.

These correspond to cases where highly similar copies of members of a family were separated from their paralogs during the clustering or assembly steps. Finally, a number of alignments had only one reference sequence from the CL Brener hybrid. These cases may correspond to haploid regions in the hybrid genome or to cases where two highly divergent alleles were separated during the clus tering step. We then scanned the multiple sequence alignments and identified columns containing sequence differ ences and or indels. From the set of all alignments we identified 325,355 sites with variation, of which 28,316 corresponded to small indels. These polymorphic sites provide representative infor mation on the diversity found in T.

cruzi evolutionary lineages TcI, TcVI, but also in lineages TcII and TcIII. Columns containing variation in a multiple sequence alignment may correspond to polymorphic sites or to sequencing errors. To discriminate between these possi bilities, we also analyzed the sequence neighborhood around each potential SNP. Based on this analysis we found 302,390 SNPs located in regions with a low density of SNPs. To further assess the quality of the sequence around in each SNP we used a statistical software package together with quality values for each base that were derived from the expected error