Maria Littmann

Maria Littmann

Auto-generated profile, based on publicly available information

Maria Littmann

Technical University of Munich

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 January 2021 in Scientific Reports
Verified
Journal Article
Open access
PDF file
MLMaria Littmann
Maria Littmann
MHMichael Heinzinger
Michael Heinzinger
CDChristian Dallago
Christian Dallago
TOTobias Olenyi
Tobias Olenyi
BRBurkhard Rost
Burkhard Rost
Knowing protein function is crucial to advance molecular and medical biology, yet experimental function annotations through the Gene Ontology (GO) exist for fewer than 0.5% of all known proteins. Computational methods bridge this sequence-annotation gap typically through homology-based annotation transfer by identifying sequence-similar proteins with known function or through prediction methods using evolutionary information. Here, we propose predicting GO terms through annotation transfer based on proximity of proteins in the SeqVec embedding rather than in sequence space. These embeddings originate from deep learned language models (LMs) for protein sequences (SeqVec) transferring the knowledge gained from predicting the next amino acid in 33 million protein sequences. Replicating the conditions of CAFA3, our method reaches an Fmax of 37 ± 2%, 50 ± 3%, and 57 ± 2% for BPO, MFO, and CCO, respectively. Numerically, this appears close to the top ten CAFA3 methods. When restricting the annotation transfer to proteins with < 20% pairwise sequence identity to the query, performance drops (Fmax BPO 33 ± 2%, MFO 43 ± 3%, CCO 53 ± 2%); this still outperforms naïve sequence-based transfer. Preliminary results from CAFA4 appear to confirm these findings. Overall, this new concept is likely to change the annotation of proteins, in particular for proteins from smaller families or proteins with intrinsically disordered regions.

13 December 2021 in Scientific Reports
Verified
Journal Article
Open access
PDF file
MLMaria Littmann
Maria Littmann
MHMichael Heinzinger
Michael Heinzinger
CDChristian Dallago
Christian Dallago
KWKonstantin Weissenow
Konstantin Weissenow
BRBurkhard Rost
Burkhard Rost
One important aspect of protein function is the binding of proteins to ligands, including small molecules, metal ions, and macromolecules such as DNA or RNA. Despite decades of experimental progress many binding sites remain obscure. Here, we proposed bindEmbed21, a method predicting whether a protein residue binds to metal ions, nucleic acids, or small molecules. The Artificial Intelligence (AI)-based method exclusively uses embeddings from the Transformer-based protein Language Model (pLM) ProtT5 as input. Using only single sequences without creating multiple sequence alignments (MSAs), bindEmbed21DL outperformed MSA-based predictions. Combination with homology-based inference increased performance to F1 = 48 ± 3% (95% CI) and MCC = 0.46 ± 0.04 when merging all three ligand classes into one. All results were confirmed by three independent data sets. Focusing on very reliably predicted residues could complement experimental evidence: For the 25% most strongly predicted binding residues, at least 73% were correctly predicted even when ignoring the problem of missing experimental annotations. The new method bindEmbed21 is fast, simple, and broadly applicable—neither using structure nor MSAs. Thereby, it found binding residues in over 42% of all human proteins not otherwise implied in binding and predicted about 6% of all residues as binding to metal ions, nucleic acids, or small molecules.

11 June 2022 in NAR Genomics and Bioinformatics
Verified
Journal Article
Open access
PDF file
MHMichael Heinzinger
Michael Heinzinger
MLMaria Littmann
Maria Littmann
ISIan Sillitoe
Ian Sillitoe
NBNicola Bordin
Nicola Bordin
COChristine Orengo
Christine Orengo
BRBurkhard Rost
Burkhard Rost
Experimental structures are leveraged through multiple sequence alignments, or more generally through homology-based inference (HBI), facilitating the transfer of information from a protein with known annotation to a query without any annotation. A recent alternative expands the concept of HBI from sequence-distance lookup to embedding-based annotation transfer (EAT). These embeddings are derived from protein Language Models (pLMs). Here, we introduce using single protein representations from pLMs for contrastive learning. This learning procedure creates a new set of embeddings that optimizes constraints captured by hierarchical classifications of protein 3D structures defined by the CATH resource. The approach, dubbed ProtTucker, has an improved ability to recognize distant homologous relationships than more traditional techniques such as threading or fold recognition. Thus, these embeddings have allowed sequence comparison to step into the ‘midnight zone’ of protein similarity, i.e. the region in which distantly related sequences have a seemingly random pairwise sequence similarity. The novelty of this work is in the particular combination of tools and sampling techniques that ascertained good performance comparable or better to existing state-of-the-art sequence comparison methods. Additionally, since this method does not need to generate alignments it is also orders of magnitudes faster. The code is available at https://github.com/Rostlab/EAT.

9 December 2022 in Trends in Biochemical Sciences
Verified
Journal Article
Open access
NBNicola Bordin
Nicola Bordin
CDChristian Dallago
Christian Dallago
MHMichael Heinzinger
Michael Heinzinger
SKStephanie Kim
Stephanie Kim
MLMaria Littmann
Maria Littmann
COChristine Orengo
Christine Orengo
Breakthrough methods in machine learning (ML), protein structure prediction, and novel ultrafast structural aligners are revolutionizing structural biology. Obtaining accurate models of proteins and annotating their functions on a large scale is no longer limited by time and resources. The most recent method to be top ranked by the Critical Assessment of Structure Prediction (CASP) assessment, AlphaFold 2 (AF2), is capable of building structural models with an accuracy comparable to that of experimental structures. Annotations of 3D models are keeping pace with the deposition of the structures due to advancements in protein language models (pLMs) and structural aligners that help validate these transferred annotations. In this review we describe how recent developments in ML for protein science are making large-scale structural bioinformatics available to the general scientific community.

13 January 2020 in Nature Machine Intelligence
Verified
Journal Article
Closed access
PDF file
MLMaria Littmann
Maria Littmann
KSKatharina Selig
Katharina Selig
LCLiel Cohen-Lavi
Liel Cohen-Lavi
YFYotam Frank
Yotam Frank
PHPeter Hönigschmid
Peter Hönigschmid
BRBurkhard Rost
Burkhard Rost

6 October 2025 in Nature Medicine
Verified
Journal Article
Open access
VPVitaly Polisky
Vitaly Polisky
MLMaria Littmann
Maria Littmann
ATAleksei Triastcyn
Aleksei Triastcyn
MHMax Horn
Max Horn
AGAndreas Georgiou
Andreas Georgiou
PSPatrick Schwab
Patrick Schwab
Varicella-zoster virus (VZV) is a neurotropic virus that establishes lifelong latency in humans. VZV reactivation is associated with a wide range of symptoms, including herpes zoster (HZ; also known as shingles), and has been implicated in the development of dementia, although to an unknown extent. Here we present a large-scale longitudinal analysis of health records from more than 100 million individuals in the United States that demonstrates a consistent relationship of VZV reactivation with dementia after controlling for nearly 400 measured characteristics (covering demographics, socioeconomic factors, comorbidities, medications, proxies for healthcare-seeking behavior, shifts in clinical guidelines and completeness of records). We found that recurrent HZ was associated with an increased risk of dementia compared to a single HZ episode. Additionally, exposure to HZ vaccines was associated with a reduced risk of dementia compared to the control 23-valent pneumococcal polysaccharide vaccine. Furthermore, the reduced risk of dementia after administration of the live-attenuated zoster vaccine waned over time and was highly correlated with a waning of the vaccine-mediated protection against HZ. The dementia risk reduction at 3 and 5 years postexposure was also stronger in individuals who received multiple as opposed to only one dose of the recombinant HZ vaccine and those at greater risk of HZ. Our findings strongly implicate VZV reactivation as a modifiable risk factor for dementia.