view release on metacpan or  search on metacpan

lib/AI/TensorFlow/Libtensorflow/Manual/Notebook/InferenceUsingTFHubEnformerGeneExprPredModel.pod  view on Meta::CPAN

  PDL: Float D [5313,896]

=head2 Plot predicted tracks

These predictions can be plotted 

  my @tracks = (
      [ 'DNASE:CD14-positive monocyte female' =>   41 => $predictions_p->slice('(41)') ],
      [ 'DNASE:keratinocyte female'           =>   42 => $predictions_p->slice('(42)') ],
      [ 'CHIP:H3K27ac:keratinocyte female'    =>  706 => $predictions_p->slice('(706)')],
      [ 'CAGE:Keratinocyte - epidermal'       => 4799 => log10(1 + $predictions_p->slice('(4799)')) ],
  );
  
  use PDL::Graphics::Gnuplot;
  
  my $plot_output_path = 'enformer-target-interval-tracks.png';
  my $gp = gpwin('pngcairo', font => ",10", output => $plot_output_path, size => [10,2. * @tracks], aa => 2 );
  
  $gp->multiplot( layout => [1, scalar @tracks], title => $target_interval );
  
  $gp->options(
      offsets => [ graph => "0.01, 0, 0, 0" ],
      lmargin => "at screen 0.05",
  );
  
  my $x = zeroes($predictions_p->dim(1))->xlinvals($target_interval->start, $target_interval->end);
  
  my @tics_opts = (mirror => 0, out => 1);
  
  for my $i (0..$#tracks) {
      my ($title, $id, $y) = @{$tracks[$i]};
      $gp->plot( {
              title => $title,
              border => [2],
              ytics => { @tics_opts, locations => [ ceil(($y->max-$y->min)/2)->sclr ] },
              ( $i == $#tracks
                  ? ( xtics => { format => '%.3f', @tics_opts } )
                  : ( xtics => 0 ) ),
              ( $i == $#tracks ? ( xlabel => 'location ({/Symbol \264}10^7 bases)' ) : ()  ),
  
          },
          with => 'filledcurves',
          #'lc' => '#086eb5',
  
          # $x scaled by 1e7; filled curve between $y and the x-axis
          $x / 1e7, $y, pdl(0)
      );
  }
  
  $gp->end_multi;
  
  $gp->close;
  
  if( IN_IPERL ) {
      IPerl->png( bytestream => path($plot_output_path)->slurp_raw );
  }

B<DISPLAY>:

=for html <span style="display:inline-block;margin-left:1em;"><p><img						src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA+gAAAMgCAIAAAA/et9qAAAgAElEQVR4nOzdd2AUVeIH8Ddb0jshBAIEpSo1GjoIpyAgCOqd3uGdoGBBUQQFRUVBRbkTf9gOBQucqFiwUhSSgJQYCCSBkJBAet1k...

=head2 Parts of the original notebook that fall outside the scope

In the orignal notebook, there are several more steps that have not been ported here:

=over

=item 1.

"Compute contribution scores":

This task requires implementing C<@tf.function> to compile gradients.

=item 2.

"Predict the effect of a genetic variant" and "Score multiple variants":

The first task is possible, but the second task requires loading a pre-processing pipeline for scikit-learn and unfortunately this pipeline is stored as a pickle file that is valid for an older version of scikit-learn (version 0.23.2) and as such its...

=back

  # Some code that could be used for working with variants.
  1 if <<'COMMENT';
  
  use Bio::DB::HTS::VCF;
  
  my $clinvar_tbi_path = "${clinvar_path}.tbi";
  unless( -f $clinvar_tbi_path ) {
      system( qw(tabix), $clinvar_path );
  }
  my $v = Bio::DB::HTS::VCF->new( filename => $clinvar_path );
  $v->num_variants
  
  COMMENT
  
  undef;

=head1 RESOURCE USAGE

  use Filesys::DiskUsage qw/du/;
  
  my $total = du( { 'human-readable' => 1, dereference => 1 },
      $model_archive_path, $model_base, $new_model_base,
  
      $targets_path,
  
      $hg_gz_path,
      $hg_bgz_path, $hg_bgz_fai_path,
  
      $clinvar_path,
  
      $plot_output_path,
  );
  
  say "Disk space usage: $total"; undef;

B<STREAM (STDOUT)>:

  Disk space usage: 4.66G