The columns represent:

first - chromosome name

second and third - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth - average expression (reads per million, RPM), calculated as the average of within-tissue mean expression values, to avoid biases from variable number of samples in different tissues. Note that the meaning of this column changed since PolyASite v2.0, where each sample contributed equally to the average.

sixth - strand on which the PAS is encoded

seventh - percentage of tissues that support the particular PAS

eighth - number of different experimental protocols that support the particular PAS (in this case 1, 10X genomics-based scRNA-seq)

ninth - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

tenth - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0. Specifically, in v2.0 each PAS could be assigned to several genes, including in anti-sense direction, and alternatively-spliced (exonic/intronic) regions were not distinguished as a separate class. In v3.0, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand.

eleventh - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

The columns represent:

first (chrom) - chromosome name

second (chromStart) and third (chromEnd) - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth (name) - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth (avg_expression) - average expression (reads per million, RPM), calculated as the average of within-tissue mean expression values, to avoid biases from variable number of samples in different tissues. Note that the meaning of this column changed since PolyASite v2.0, where each sample contributed equally to the average.

sixth (strand) - strand on which the PAS is encoded

seventh (rep) - representative cleavage site of the cluster (genomic coordinate)

eighth (perc_tissues) - percentage of tissues that support the particular PAS

ninth (nr_prots) - number of different experimental protocols that support the particular PAS

tenth (annotation) - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0 where each PAS could be assigned to several genes, including in anti-sense direction, and alternative (exonic/intronic) regions were not distinguished as a separate class. Now, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or the region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest 3’end on the same strand.

eleventh (gene_name) - gene symbol of the annotated gene to which the PAS is assigned. Note that each PAS is now uniquely assigned to one gene. In case the PAS coordinate is located in an intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand. This classification scheme is different from that of PolyASite v2.0, where a PAS could be assigned to multiple overlapping genes. Moreover PAS were associated with genes only when the genomic coordinates of PAS directly overlapped with those of the gene.

twelfth (gene_id) - gene id of the annotated gene to which PAS is assigned. See the description for column 11 for more details.

thirteenth (repSite_signals) - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

fourteenth (stringency_level) - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

fifteenth onwards - average expression (reads per million, RPM) within tissues. Note that the meaning of these columns changed since PolyASite v2.0, where these columns represented RPM values within each of the analyzed samples.

The columns represent:

first - chromosome name

second and third - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth - average expression (reads per million, RPM), calculated as the average of within-tissue mean expression values, to avoid biases from variable number of samples in different tissues. Note that the meaning of this column changed since PolyASite v2.0, where each sample contributed equally to the average.

sixth - strand on which the PAS is encoded

seventh - percentage of tissues that support the particular PAS

eighth - number of different experimental protocols that support the particular PAS (in this case 1, 10X genomics-based scRNA-seq)

ninth - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

tenth - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0. Specifically, in v2.0 each PAS could be assigned to several genes, including in anti-sense direction, and alternatively-spliced (exonic/intronic) regions were not distinguished as a separate class. In v3.0, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand.

eleventh - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

The columns represent:

first (chrom) - chromosome name

second (chromStart) and third (chromEnd) - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth (name) - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth (avg_expression) - average expression (reads per million, RPM), calculated as the average of within-tissue mean expression values, to avoid biases from variable number of samples in different tissues. Note that the meaning of this column changed since PolyASite v2.0, where each sample contributed equally to the average.

sixth (strand) - strand on which the PAS is encoded

seventh (rep) - representative cleavage site of the cluster (genomic coordinate)

eighth (perc_tissues) - percentage of tissues that support the particular PAS

ninth (nr_prots) - number of different experimental protocols that support the particular PAS

tenth (annotation) - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0 where each PAS could be assigned to several genes, including in anti-sense direction, and alternative (exonic/intronic) regions were not distinguished as a separate class. Now, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or the region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest 3’end on the same strand.

eleventh (gene_name) - gene symbol of the annotated gene to which the PAS is assigned. Note that each PAS is now uniquely assigned to one gene. In case the PAS coordinate is located in an intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand. This classification scheme is different from that of PolyASite v2.0, where a PAS could be assigned to multiple overlapping genes. Moreover PAS were associated with genes only when the genomic coordinates of PAS directly overlapped with those of the gene.

twelfth (gene_id) - gene id of the annotated gene to which PAS is assigned. See the description for column 11 for more details.

thirteenth (repSite_signals) - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

fourteenth (stringency_level) - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

fifteenth onwards - average expression (reads per million, RPM) within tissues. Note that the meaning of these columns changed since PolyASite v2.0, where these columns represented RPM values within each of the analyzed samples.

The columns represent:

first - chromosome name

second and third - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth - average expression (reads per million, RPM) over samples

sixth - strand on which the PAS is encoded

seventh - percentage of samples that support the particular PAS

eighth - number of different experimental protocols that support the particular PAS (in this case 1, 10X genomics-based scRNA-seq)

ninth - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

tenth - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0. Specifically, in v2.0 each PAS could be assigned to several genes, including in anti-sense direction, and alternatively-spliced (exonic/intronic) regions were not distinguished as a separate class. In v3.0, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand.

eleventh - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

The columns represent:

first (chrom) - chromosome name

second (chromStart) and third (chromEnd) - start and end positions of the PAS, respectively. Differences > 1 between start and end positions reflect the imprecision of the cleavage process.

fourth (name) - unique PAS ID, composed of the chromosome name, the coordinate of the representative cleavage site of the PAS and the strand. Note that this ID format is inspired by that of UCSC's genome browser and uses 1-based instead of 0-based coordinates, as per bed file format. Thus, to convert the coordinate of the representative site to bed coordinates, subtract 1 from the given coordinate of the representative site to have a start coordinate.

fifth (avg_expression) - average expression (reads per million, RPM) over samples

sixth (strand) - strand on which the PAS is encoded

seventh (rep) - representative cleavage site of the cluster (genomic coordinate)

eighth (perc_samples) - percentage of samples that support the particular PAS

ninth (nr_prots) - number of different experimental protocols that support the particular PAS

tenth (annotation) - two letter code for the genomic class annotation of the PAS (TE, terminal exon; EX, exonic; IN, intronic; AL, alternative (exonic/intronic); DI, downstream intergenic; UI, upstream intergenic; NA, not available (ambiguous)). See Figure 1C of the PolyASite v3.0 paper for details. Note that the classification scheme changed since PolyASite v2.0 where each PAS could be assigned to several genes, including in anti-sense direction, and alternative (exonic/intronic) regions were not distinguished as a separate class. Now, each PAS is uniquely assigned to one gene. In case the PAS coordinate is located in the intergenic region or the region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest 3’end on the same strand.

eleventh (gene_name) - gene symbol of the annotated gene to which the PAS is assigned. Note that each PAS is now uniquely assigned to one gene. In case the PAS coordinate is located in an intergenic region or in a region where multiple genes overlap on the same strand, PAS is assigned to the gene with the closest annotated 3’end on the same strand. This classification scheme is different from that of PolyASite v2.0, where a PAS could be assigned to multiple overlapping genes. Moreover PAS were associated with genes only when the genomic coordinates of PAS directly overlapped with those of the gene.

twelfth (gene_id) - gene id of the annotated gene to which PAS is assigned. See the description for column 11 for more details.

thirteenth (repSite_signals) - information about the poly(A) signal(s) (core hexamer) that are present in the region from 35 to 10 nts upstream of the PAS. Note that in PolyASite v2.0, a larger window of 60 nt upstream to 10 nt downstream relative to the cleavage site, was used, and 18 motifs were considered, instead of 12 as in PolyASite v3.0.

fourteenth (stringency_level) - stringency level, the maximal motif percentage at which the PAS is retained upon filtration. See Figure 1B and Supplementary Figure 1 of the PolyASite v3.0 paper for details. Note that the meaning of this column changed since PolyASite v2.0, where it contained the average expression, as column 5.

fifteenth onwards - average expression (reads per million, RPM) within samples