David Calderón-Franco
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
David Calderón-Franco has a total of 70 co-authors. Here are the top co-authors by number of shared publications:
- David Gregory Weissbrodt
Department of Biotechnology, Delft University of Technology, Delft, Netherlands
16 shared publications - Mark Cornelis Maria van Loosdrecht
Department of Biotechnology, Delft University of Technology, van der Maasweg 9, 2629 HZ Delft, The Netherlands
11 shared publications - Thomas Abeel
Delft University of Technology
7 shared publications - Antonella L. Piaggio
Faculty of Civil Engineering and Geosciences, Section Sanitary Engineering, Department of Water Management, Delft University of Technology, Stevinweg 1, Delft, South Holland 2628 CN, The Netherlands
2 shared publications Ben Abbas
Delft University of Technology
2 shared publications- Bianca-Maria Cosma
Delft Bioinformatics Lab, Delft University of Technology, Delft, The Netherlands
2 shared publications
Latest Publications
- Open Access
- 10 April 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (4) , e0345919
Predicted meta-omics: A potential solution to multi-omics data scarcity in microbiome studies
Imbalances in the gut microbiome have been linked to conditions such as inflammatory bowel disease, diabetes, and cancer. While metagenomics and amplicon sequencing are commonly used to study the microbiome, they do not capture all layers of microbial functions. Other meta-omics data can provide more insights, but these are more costly and laborious to procure. The growing availability of paired meta-omics data offers an opportunity to develop machine learning models that can infer connections between metagenomics data and other forms of meta-omics data, enabling the prediction of these other forms of meta-omics data from metagenomics. We evaluated several machine learning models for predicting meta-omics features from various meta-omics inputs. Simpler architectures such as elastic net regression and random forests generated reliable predictions of transcript and metabolite abundances, with correlations of up to 0.77 and 0.74, respectively, but predicting protein profiles was more challenging. We also identified a core set of well-predicted features for each meta-omics output type, and showed that multi-output regression neural networks performed similarly when trained using fewer output features. Lastly, our experiments demonstrated that predicted features can be used for the downstream task of inflammatory bowel disease classification, with performance comparable to that of experimental data.
- 10 April 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (4) , e0345919
Open AccessPredicted meta-omics: A potential solution to multi-omics data scarcity in microbiome studies
Imbalances in the gut microbiome have been linked to conditions such as inflammatory bowel disease, diabetes, and cancer. While metagenomics and amplicon sequencing are commonly used to study the microbiome, they do not capture all layers of microbial functions. Other meta-omics data can provide more insights, but these are more costly and laborious to procure. The growing availability of paired meta-omics data offers an opportunity to develop machine learning models that can infer connections between metagenomics data and other forms of meta-omics data, enabling the prediction of these other forms of meta-omics data from metagenomics. We evaluated several machine learning models for predicting meta-omics features from various meta-omics inputs. Simpler architectures such as elastic net regression and random forests generated reliable predictions of transcript and metabolite abundances, with correlations of up to 0.77 and 0.74, respectively, but predicting protein profiles was more challenging. We also identified a core set of well-predicted features for each meta-omics output type, and showed that multi-output regression neural networks performed similarly when trained using fewer output features. Lastly, our experiments demonstrated that predicted features can be used for the downstream task of inflammatory bowel disease classification, with performance comparable to that of experimental data.
- Open Access
Predicted meta-omics: a potential solution to multi-omics data scarcity in microbiome studies
Imbalances in the gut microbiome have been linked to conditions such as inflammatory bowel disease, diabetes, and cancer. While metagenomics and amplicon sequencing are commonly used to study the microbiome, they do not capture all layers of microbial functions. Other meta-omics data can provide more insights, but these are more costly and laborious to procure. The growing availability of paired meta-omics data offers an opportunity to develop machine learning models that can infer connections between metagenomics data and other forms of meta-omics data, enabling the prediction of these other forms of meta-omics data from metagenomics. We evaluated several machine learning models for predicting meta-omics features from various meta-omics inputs. Simpler architectures such as elastic net regression and random forests generated reliable predictions of transcript and metabolite abundances, with correlations of up to 0.77 and 0.74, respectively, but predicting protein profiles was more challenging. We also identified a core set of well-predicted features for each meta-omics output type, and showed that multi-output regression neural networks performed similarly when trained using fewer output features. Lastly, our experiments demonstrated that predicted features can be used for the downstream task of inflammatory bowel disease classification, with performance comparable to that of experimental data.
Open AccessPredicted meta-omics: a potential solution to multi-omics data scarcity in microbiome studies
Imbalances in the gut microbiome have been linked to conditions such as inflammatory bowel disease, diabetes, and cancer. While metagenomics and amplicon sequencing are commonly used to study the microbiome, they do not capture all layers of microbial functions. Other meta-omics data can provide more insights, but these are more costly and laborious to procure. The growing availability of paired meta-omics data offers an opportunity to develop machine learning models that can infer connections between metagenomics data and other forms of meta-omics data, enabling the prediction of these other forms of meta-omics data from metagenomics. We evaluated several machine learning models for predicting meta-omics features from various meta-omics inputs. Simpler architectures such as elastic net regression and random forests generated reliable predictions of transcript and metabolite abundances, with correlations of up to 0.77 and 0.74, respectively, but predicting protein profiles was more challenging. We also identified a core set of well-predicted features for each meta-omics output type, and showed that multi-output regression neural networks performed similarly when trained using fewer output features. Lastly, our experiments demonstrated that predicted features can be used for the downstream task of inflammatory bowel disease classification, with performance comparable to that of experimental data.
- Open Access
- 1 May 2024
- Published by Royal Society of Chemistry (RSC) in Environmental Science: Water Research & Technology
- Vol. 10 (6) , 1406-1420
Antibiotic resistance response of activated sludge to sulfamethoxazole: insights from the intracellular and extracellular DNA fractions
In activated sludge, the antibiotic resistance genes (ARGs) can be present either in the intracellular (iDNA) or extracellular DNA fraction (exDNA). Recent advances in the exDNA extraction methodology allow a better profiling of the pool of ARGs. However, little is known about how stress conditions modify the distribution of ARGs between both DNA fractions. Here, we performed two batch tests for analyzing the effects of two different stress conditions, namely nutrient starvation and high concentrations of sulfamethoxazole (1, 10 and 150 mg L−1) in activated sludge. We tracked by qPCR the resulting relative abundances of four target genes, namely the universal 16S rRNA gene, the class 1 integron-integrase gene intI1, and the sulfonamide resistance genes sul1 and sul2 in both the iDNA and exDNA fractions. In the exDNA pool, unlike starvation, which provoked a decrease of 1–2 log10 [copies] per ng DNA in the concentration of sul1 and intI1, the presence of sulfamethoxazole did not influence the abundances of sul1 and sul2. However, high concentrations of sulfamethoxazole (150 mg L−1) selected for microorganisms harboring sul1 and, more remarkably, sul2 genes in their iDNA during their exponential growth phase. The abundances of intI1 and sul1 were positively correlated in the exDNA fraction (r > 0.7), whereas no significant correlation (p < 0.05) between the abundance of these two genes was found in the iDNA fraction of the sludge. High SMX concentrations influenced the abundance of ARGs in the iDNA; their abundance in the exDNA was influenced by nutrient limitations. Further studies should consider the profiling of exDNA fractions because of the relationship between ARGs and mobile genetic elements. Besides, the surveillance of antimicrobial resistance is encouraged in wastewater treatment plants facing high antibiotic concentrations.
- 1 May 2024
- Published by Royal Society of Chemistry (RSC) in Environmental Science: Water Research & Technology
- Vol. 10 (6) , 1406-1420
Open AccessAntibiotic resistance response of activated sludge to sulfamethoxazole: insights from the intracellular and extracellular DNA fractions
In activated sludge, the antibiotic resistance genes (ARGs) can be present either in the intracellular (iDNA) or extracellular DNA fraction (exDNA). Recent advances in the exDNA extraction methodology allow a better profiling of the pool of ARGs. However, little is known about how stress conditions modify the distribution of ARGs between both DNA fractions. Here, we performed two batch tests for analyzing the effects of two different stress conditions, namely nutrient starvation and high concentrations of sulfamethoxazole (1, 10 and 150 mg L−1) in activated sludge. We tracked by qPCR the resulting relative abundances of four target genes, namely the universal 16S rRNA gene, the class 1 integron-integrase gene intI1, and the sulfonamide resistance genes sul1 and sul2 in both the iDNA and exDNA fractions. In the exDNA pool, unlike starvation, which provoked a decrease of 1–2 log10 [copies] per ng DNA in the concentration of sul1 and intI1, the presence of sulfamethoxazole did not influence the abundances of sul1 and sul2. However, high concentrations of sulfamethoxazole (150 mg L−1) selected for microorganisms harboring sul1 and, more remarkably, sul2 genes in their iDNA during their exponential growth phase. The abundances of intI1 and sul1 were positively correlated in the exDNA fraction (r > 0.7), whereas no significant correlation (p < 0.05) between the abundance of these two genes was found in the iDNA fraction of the sludge. High SMX concentrations influenced the abundance of ARGs in the iDNA; their abundance in the exDNA was influenced by nutrient limitations. Further studies should consider the profiling of exDNA fractions because of the relationship between ARGs and mobile genetic elements. Besides, the surveillance of antimicrobial resistance is encouraged in wastewater treatment plants facing high antibiotic concentrations.
- Open Access
- 19 October 2023
- Published by Elsevier in Water Research
- Vol. 247, 120761
Candidate biomarkers of antibiotic resistance for the monitoring of wastewater and the downstream environment
- 19 October 2023
- Published by Elsevier in Water Research
- Vol. 247, 120761
Open AccessCandidate biomarkers of antibiotic resistance for the monitoring of wastewater and the downstream environment
- Open Access
- 11 October 2023
- Published by IWA Publishing in Water Science & Technology
- Vol. 88 (9) , 2344-2363
The fate of sulfamethoxazole and trimethoprim in a micro-aerated anaerobic membrane bioreactor and the occurrence of antibiotic resistance in the permeate
This study investigates the effects, conversions, and resistance induction, following the addition of 150 μg·L−1 of two antibiotics, sulfamethoxazole (SMX) and trimethoprim (TMP), in a laboratory-scale micro-aerated anaerobic membrane bioreactor (MA-AnMBR). TMP and SMX were removed at 97 and 86%, indicating that micro-aeration did not hamper their removal. These antibiotics only affected the pH and biogas composition of the process, with a significant change in pH from 7.8 to 7.5, and a decrease in biogas methane content from 84 to 78%. TMP was rapidly adsorbed onto the sludge and subsequently degraded during the long solids retention time of 27 days. SMX adsorption was minimal, but the applied hydraulic retention time of 2.6 days was sufficiently long to biodegrade SMX. The levels of three antibiotic-resistant genes (ARGs) (sul1, sul2, and dfrA1) and one mobile genetic element biomarker (intI1) were analyzed by qPCR. Additions of the antibiotics increased the relative abundances of all ARGs and intI1 in the MA-AnMBR sludge, with the sul2 gene folding 15 times after 310 days of operation. The MA-AnMBR was able to reduce the concentration of antibiotic-resistant bacteria (ARB) in the permeate by 3 log.
- 11 October 2023
- Published by IWA Publishing in Water Science & Technology
- Vol. 88 (9) , 2344-2363
Open AccessThe fate of sulfamethoxazole and trimethoprim in a micro-aerated anaerobic membrane bioreactor and the occurrence of antibiotic resistance in the permeate
This study investigates the effects, conversions, and resistance induction, following the addition of 150 μg·L−1 of two antibiotics, sulfamethoxazole (SMX) and trimethoprim (TMP), in a laboratory-scale micro-aerated anaerobic membrane bioreactor (MA-AnMBR). TMP and SMX were removed at 97 and 86%, indicating that micro-aeration did not hamper their removal. These antibiotics only affected the pH and biogas composition of the process, with a significant change in pH from 7.8 to 7.5, and a decrease in biogas methane content from 84 to 78%. TMP was rapidly adsorbed onto the sludge and subsequently degraded during the long solids retention time of 27 days. SMX adsorption was minimal, but the applied hydraulic retention time of 2.6 days was sufficiently long to biodegrade SMX. The levels of three antibiotic-resistant genes (ARGs) (sul1, sul2, and dfrA1) and one mobile genetic element biomarker (intI1) were analyzed by qPCR. Additions of the antibiotics increased the relative abundances of all ARGs and intI1 in the MA-AnMBR sludge, with the sul2 gene folding 15 times after 310 days of operation. The MA-AnMBR was able to reduce the concentration of antibiotic-resistant bacteria (ARB) in the permeate by 3 log.
- Open Access
The fate of sulfamethoxazole and trimethoprim in a micro-aerated anaerobic membrane bioreactor: implications for antibiotic resistance spreading
Interest in reusing treated wastewater drives efforts to eliminate antibiotics from water sources to prevent antibiotic resistance. Micro-aerated anaerobic membrane bioreactors (MA-AnMBR) promote wastewater reuse with high organic matter conversion to biogas, under a small footprint. However, the fates of antibiotics, antibiotic-resistant bacteria (ARB), and their antibiotic-resistance genes (ARGs) are not known in these systems. We studied the effects, conversions, and resistance induction, following the addition of 150 μg·L -1 of two antibiotics, sulfamethoxazole (SMX) and trimethoprim (TMP), in a laboratory-scale MA-AnMBR. TMP and SMX were removed at 97 and 86%, indicating that micro-aeration did not hamper the removal of the antibiotics. These antibiotics only affected the pH and biogas composition of the process, with a significant change in pH from 7.8 to 7.5, and a decrease in biogas CH 4 content from 84 to 78%. TMP was rapidly adsorbed onto the sludge and subsequently degraded during the long retention of the solids of 27 days. SMX adsorption was minimal, but the applied hydraulic retention time of 2.6 days was sufficiently long to biodegrade SMX. The levels of three ARGs ( sul1 and sul2 for SMX, dfrA1 ) and one mobile genetic element biomarker ( intI1 ) were analysed by qPCR, in combination with ARB tracked by plating. Additions of the antibiotics increased the relative abundances of all ARGs and intI1 in the MA-AnMBR sludge, with the sul2 gene folding 15 times after 310 days of operation. The MA-AnMBR was able to reduce the concentration of ARB in the permeate by 3 log. Highlights: Additions of SMX and TMP had a negligible effect on the MA-AnMBR performance. The laboratory-scale MA-AnMBR removed 86% of SMX and 97% of TMP. A 3 log removal of ARB was achieved between sludge and UF permeate. Relative abundances of ARGs were similar in sludge and permeate. TMP and SMX resistance is better assessed by the heterotrophic plate count of ARB.
Open AccessThe fate of sulfamethoxazole and trimethoprim in a micro-aerated anaerobic membrane bioreactor: implications for antibiotic resistance spreading
Interest in reusing treated wastewater drives efforts to eliminate antibiotics from water sources to prevent antibiotic resistance. Micro-aerated anaerobic membrane bioreactors (MA-AnMBR) promote wastewater reuse with high organic matter conversion to biogas, under a small footprint. However, the fates of antibiotics, antibiotic-resistant bacteria (ARB), and their antibiotic-resistance genes (ARGs) are not known in these systems. We studied the effects, conversions, and resistance induction, following the addition of 150 μg·L -1 of two antibiotics, sulfamethoxazole (SMX) and trimethoprim (TMP), in a laboratory-scale MA-AnMBR. TMP and SMX were removed at 97 and 86%, indicating that micro-aeration did not hamper the removal of the antibiotics. These antibiotics only affected the pH and biogas composition of the process, with a significant change in pH from 7.8 to 7.5, and a decrease in biogas CH 4 content from 84 to 78%. TMP was rapidly adsorbed onto the sludge and subsequently degraded during the long retention of the solids of 27 days. SMX adsorption was minimal, but the applied hydraulic retention time of 2.6 days was sufficiently long to biodegrade SMX. The levels of three ARGs ( sul1 and sul2 for SMX, dfrA1 ) and one mobile genetic element biomarker ( intI1 ) were analysed by qPCR, in combination with ARB tracked by plating. Additions of the antibiotics increased the relative abundances of all ARGs and intI1 in the MA-AnMBR sludge, with the sul2 gene folding 15 times after 310 days of operation. The MA-AnMBR was able to reduce the concentration of ARB in the permeate by 3 log. Highlights: Additions of SMX and TMP had a negligible effect on the MA-AnMBR performance. The laboratory-scale MA-AnMBR removed 86% of SMX and 97% of TMP. A 3 log removal of ARB was achieved between sludge and UF permeate. Relative abundances of ARGs were similar in sludge and permeate. TMP and SMX resistance is better assessed by the heterotrophic plate count of ARB.
- Open Access
Influence of DNA extraction methods on microbiome and resistome analysis in activated sludge
Amplicon sequencing, metagenomics, and quantitative polymerase chain reaction (qPCR) are commonly used techniques to analyse microorganisms and antibiotic resistance genes (ARGs) in activated sludge from wastewater treatment plants (WWTPs). However, the lack of workflow harmonisation poses challenges in comparing measurements across studies and research groups. To address this issue, we examined the impact of DNA extraction procedures on 16S rRNA gene amplicon sequencing, shotgun metagenomics, and qPCR analyses of activated sludge by combining two widely used DNA extraction kits (PowerSoil and FastDNA) and two commonly employed disruption instruments (bead-beater and vortex) through a 2×2 factorial experimental design involving four groups of three analysts performing DNA extractions in triplicates. Our findings revealed significant differences in DNA yield, purity, and reproducibility of amplicon sequencing profiles among the extraction kits. Operator variability also influenced the results. We compared microbiome profiles obtained by amplicon sequencing and metagenomics and observed that bead-beating introduced more variability among triplicates compared to vortexing. The combinations of extraction kits and disruption instruments impacted the relative abundances of specific phyla such as Actinobacteriota , Bacteroidota , and Nitrospirota . For resistome analysis, we employed metagenomics for high-resolution profiling and qPCR for high-sensitivity detection of ARGs. The compositions and diversities of resistome datasets were not significantly affected by the choice of extraction kits and disruption instruments. Although using the same method is ideal for accurate comparisons, our results suggest that acceptable reproducibility can still be achieved when using different methods. This finding encourages the implementation of ARG monitoring in wastewater treatment processes. However, it is important to consider biases introduced by DNA extraction workflows when designing analytical studies, interpreting their results, and comparing their findings. Striving for more harmonised molecular workflows is crucial in the field of wastewater microbiology and engineering. Abstract Figure: Graphical abstract
Open AccessInfluence of DNA extraction methods on microbiome and resistome analysis in activated sludge
Amplicon sequencing, metagenomics, and quantitative polymerase chain reaction (qPCR) are commonly used techniques to analyse microorganisms and antibiotic resistance genes (ARGs) in activated sludge from wastewater treatment plants (WWTPs). However, the lack of workflow harmonisation poses challenges in comparing measurements across studies and research groups. To address this issue, we examined the impact of DNA extraction procedures on 16S rRNA gene amplicon sequencing, shotgun metagenomics, and qPCR analyses of activated sludge by combining two widely used DNA extraction kits (PowerSoil and FastDNA) and two commonly employed disruption instruments (bead-beater and vortex) through a 2×2 factorial experimental design involving four groups of three analysts performing DNA extractions in triplicates. Our findings revealed significant differences in DNA yield, purity, and reproducibility of amplicon sequencing profiles among the extraction kits. Operator variability also influenced the results. We compared microbiome profiles obtained by amplicon sequencing and metagenomics and observed that bead-beating introduced more variability among triplicates compared to vortexing. The combinations of extraction kits and disruption instruments impacted the relative abundances of specific phyla such as Actinobacteriota , Bacteroidota , and Nitrospirota . For resistome analysis, we employed metagenomics for high-resolution profiling and qPCR for high-sensitivity detection of ARGs. The compositions and diversities of resistome datasets were not significantly affected by the choice of extraction kits and disruption instruments. Although using the same method is ideal for accurate comparisons, our results suggest that acceptable reproducibility can still be achieved when using different methods. This finding encourages the implementation of ARG monitoring in wastewater treatment processes. However, it is important to consider biases introduced by DNA extraction workflows when designing analytical studies, interpreting their results, and comparing their findings. Striving for more harmonised molecular workflows is crucial in the field of wastewater microbiology and engineering. Abstract Figure: Graphical abstract
- Open Access
- 27 March 2023
- Published by Elsevier in Water Research
- Vol. 235, 119905
Microbiome, resistome and mobilome of chlorine-free drinking water treatment systems
- 27 March 2023
- Published by Elsevier in Water Research
- Vol. 235, 119905
Open AccessMicrobiome, resistome and mobilome of chlorine-free drinking water treatment systems
- Open Access
Microbiome, resistome and mobilome of chlorine-free drinking water treatment systems
Drinking water treatment plants (DWTPs) are designed to remove physical, chemical, and biological contaminants. However, until recently, the role of DWTPs in minimizing the cycling of antibiotic resistance determinants has got limited attention. In particular, the risk of selecting antibiotic-resistant bacteria (ARB) is largely overlooked in chlorine-free DWTPs where biological processes are applied. Here, we combined high-throughput quantitative PCR and metagenomics to analyze the abundance and dynamics of microbial communities, antibiotic resistance genes (ARGs), and mobile genetic elements (MGEs) across the treatment trains of two chlorine-free DWTPs involving dune-based and reservoir-based systems. The microbial diversity of the water being treated increased after all biological unit operations, namely rapid and slow sand filtration (SSF), and granular activated carbon filtration. Both DWTPs reduced the concentration of ARGs and MGEs in the water by about 2.5 log gene copies mL -1 , despite their relative increase in the disinfection sub-units (SSF in dune-based and UV treatment in reservoir-based DWTPs). The total microbial concentration was also reduced (2.5 log units), and none of the DWTPs were enriched for antibiotic resistant bacteria. Our findings highlight the effectiveness of chlorine-free DWTPs in supplying safe drinking water while reducing the concentration of antibiotic resistance determinants. To the best of our knowledge, this is the first study that monitors the presence and dynamics of antibiotic resistance determinants in chlorine-free DWTPs.
Open AccessMicrobiome, resistome and mobilome of chlorine-free drinking water treatment systems
Drinking water treatment plants (DWTPs) are designed to remove physical, chemical, and biological contaminants. However, until recently, the role of DWTPs in minimizing the cycling of antibiotic resistance determinants has got limited attention. In particular, the risk of selecting antibiotic-resistant bacteria (ARB) is largely overlooked in chlorine-free DWTPs where biological processes are applied. Here, we combined high-throughput quantitative PCR and metagenomics to analyze the abundance and dynamics of microbial communities, antibiotic resistance genes (ARGs), and mobile genetic elements (MGEs) across the treatment trains of two chlorine-free DWTPs involving dune-based and reservoir-based systems. The microbial diversity of the water being treated increased after all biological unit operations, namely rapid and slow sand filtration (SSF), and granular activated carbon filtration. Both DWTPs reduced the concentration of ARGs and MGEs in the water by about 2.5 log gene copies mL -1 , despite their relative increase in the disinfection sub-units (SSF in dune-based and UV treatment in reservoir-based DWTPs). The total microbial concentration was also reduced (2.5 log units), and none of the DWTPs were enriched for antibiotic resistant bacteria. Our findings highlight the effectiveness of chlorine-free DWTPs in supplying safe drinking water while reducing the concentration of antibiotic resistance determinants. To the best of our knowledge, this is the first study that monitors the presence and dynamics of antibiotic resistance determinants in chlorine-free DWTPs.