For over three decades, the Gabor-based <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IrisCode</i> approach has been acknowledged as the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">gold standard</i> for iris recognition, mainly due to the high entropy and binary nature of its signatures. This method is highly effective in large scale environments (e.g., national ID applications), where millions of comparisons per second are required. However, it is known that non-linear deformations in the iris texture, with fibers vanishing/appearing in response to pupil dilation/contraction, often flip the signature coefficients, being the main cause for the increase of false rejections. This paper addresses this problem, describing a customised Deep Learning (DL) framework that: 1) virtually emulates the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IrisCode</i> feature encoding phase; while also 2) detects the deformations in the iris texture that may lead to bit flipping, and autonomously adapts the filter configurations for such cases. The proposed DL architecture seamlessly integrates the Gabor kernels that extract the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IrisCode</i> and a multi-scale texture analyzer, from where the biometric signatures yield. In this sense, it can be seen as an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">adaptive encoder</i> that is fully compatible to the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IrisCode</i> approach, while increasing the permanence of the signatures. The experiments were conducted in two well known datasets (CASIA-Iris-Lamp and CASIA-Iris-Thousand) and showed a notorious decrease of the mean/standard deviation values of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">genuines</i> distribution, at expenses of only a marginal deterioration in the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">impostors</i> scores. The resulting decision environments consistently reduce the levels of false rejections with respect to the baseline for most operating levels (e.g., over 50% at 1 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e</i> <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> FAR values). The source code of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DeepGabor</i> encoder is available at: https://github.com/hugomcp/DeepGabor.