TY  - JOUR
AB  - Artificial intelligence (AI) is proliferating and developing faster than any domain scientist can adapt. To support the scientific enterprise in the Helmholtz association, a network of AI specialists has been set up to disseminate AI expertise as an infrastructure among domain scientists. As this effort exposes an evolutionary step in science organization in Germany, this article aspires to describe our setup, goals, and motivations. We comment on past experiences, current developments, and future ideas as we bring our expertise as an infrastructure closer to scientists across our organization. We hope that this offers a brief yet insightful view of our activities as well as inspiration for other science organizations.
AU  - Piraud, M.
AU  - Camero, A.*
AU  - Götz, M.*
AU  - Kesselheim, S.*
AU  - Steinbach, P.*
AU  - Weigel, T.*
C1  - 67981
C2  - 54459
CY  - 50 Hampshire St, Floor 5, Cambridge, Ma 02139 Usa
TI  - Providing AI expertise as an infrastructure in academia.
JO  - Patterns
VL  - 4
IS  - 8
PB  - Cell Press
PY  - 2023
ER  - 

TY  - JOUR
AB  - Label-efficient algorithms are of central importance for machine learning applications in many medical fields, where obtaining expert annotations is often expensive and time-consuming. Soni et al. show how contrastive learning can help build classifiers for one of the oldest and most revered methods of clinical medicine: auscultation of heart and lung sounds.
AU  - Matek, C.
C1  - 64205
C2  - 52108
TI  - More than just sound: Harnessing metadata to improve neural network classifiers for medical auscultation.
JO  - Patterns
VL  - 3
IS  - 1
PY  - 2022
ER  - 

TY  - JOUR
AB  - High-throughput drug screens in cancer cell lines test compounds at low concentrations, thereby enablingthe identification of drug-sensitivity biomarkers, while resistance biomarkers remain underexplored. Dissect-ing meaningful drug responses at high concentrations is challenging due to cytotoxicity, i.e., off-target ef-fects, thus limiting resistance biomarker discovery to frequently mutated cancer genes. To address this,we interrogate subpopulations carrying sensitivity biomarkers and consecutively investigate unexpectedlyresistant (UNRES) cell lines for unique genetic alterations that may drive resistance. By analyzing theGDSC and CTRP datasets, we find 53 and 35 UNRES cases, respectively. For 24 and 28 of them, we highlightputative resistance biomarkers. We find clinically relevant cases such as EGFRT790Mmutation in NCI-H1975or PTEN loss in NCI-H1650 cells, in lung adenocarcinoma treated with EGFR inhibitors. Interrogating the un-derpinnings of drug resistance with publicly available CRISPR phenotypic assays assists in prioritizing resis-tance drivers, offering hypotheses for drug combinations.
AU  - Ayestaran, I.
AU  - Galhoz, A.
AU  - Spiegel, E
AU  - Sidders, B.*
AU  - Dry, J.R.*
AU  - Dondelinger, F.*
AU  - Bender, A.*
AU  - McDermott, U.*
AU  - Iorio, F.*
AU  - Menden, M.P.
C1  - 60491
C2  - 49289
TI  - Identification of intrinsic drug resistance and its biomarkers in high-throughput pharmacogenomic and CRISPR screens.
JO  - Patterns
VL  - 1
IS  - 5
PY  - 2020
ER  -