Critical Assessment of Metagenome Interpretation: the second round of challenges

Evaluating metagenomic software is key for optimizing metagenome interpretation and focus of the Initiative for the Critical Assessment of Metagenome Interpretation (CAMI). The CAMI II challenge engaged the community to assess methods on realistic and complex datasets with long- and short-read sequences, created computationally from around 1,700 new and known genomes, as well as 600 new plasmids and viruses. Here we analyze 5,002 results by 76 program versions. Substantial improvements were seen in assembly, some due to long-read data. Related strains still were challenging for assembly and genome recovery through binning, as was assembly quality for the latter. Profilers markedly matured, with taxon profilers and binners excelling at higher bacterial ranks, but underperforming for viruses and Archaea. Clinical pathogen detection results revealed a need to improve reproducibility. Runtime and memory usage analyses identified efficient programs, including top performers with other metrics. The results identify challenges and guide researchers in selecting methods for analyses.

This study presents the results of the second round of the Critical Assessment of Metagenome Interpretation challenges (CAMI II), which is a community-driven effort for comprehensively benchmarking tools for metagenomics data analysis.

Keeping up with the genomes: efficient learning of our increasing knowledge of the tree of life

BACKGROUND

It is a computational challenge for current metagenomic classifiers to keep up with the pace of training data generated from genome sequencing projects, such as the exponentially-growing NCBI RefSeq bacterial genome database. When new reference sequences are added to training data, statically trained classifiers must be rerun on all data, resulting in a highly inefficient process. The rich literature of "incremental learning" addresses the need to update an existing classifier to accommodate new data without sacrificing much accuracy compared to retraining the classifier with all data.

RESULTS

We demonstrate how classification improves over time by incrementally training a classifier on progressive RefSeq snapshots and testing it on: (a) all known current genomes (as a ground truth set) and (b) a real experimental metagenomic gut sample. We demonstrate that as a classifier model's knowledge of genomes grows, classification accuracy increases. The proof-of-concept naïve Bayes implementation, when updated yearly, now runs in 1/4th of the non-incremental time with no accuracy loss.

CONCLUSIONS

It is evident that classification improves by having the most current knowledge at its disposal. Therefore, it is of utmost importance to make classifiers computationally tractable to keep up with the data deluge. The incremental learning classifier can be efficiently updated without the cost of reprocessing nor the access to the existing database and therefore save storage as well as computation resources.

[Pancreatic pseudo-cyst]

Pancreatic pseudo-cyst (PKP) represents a late complication of acute and chronic pancreatitis and of pancreatic trauma. Male sex is predominant (22/27). Average age is 43.9. PKP was most frequently associated with chronic ethylism (16), chronic pancreatitis (14) and acute pancreatitis (9), cancer (1). Ultrasonography (US) is the method used for surgical indication (parietal maturation, content) and lesion follow up. Internal drainage was used for 15 patients: 13 Jurasz cystogastroanastomy and 2 transduodenal cysto duodenostomy. External drainage was used with 9 patients. Topography was: cephalic (4), corporeal (6) corporeo caudal (13); with 4 patients topography could not be set (PKP over 15 cm). Postoperative complications occurred with 7 patients: upper digestive bleeding (4), pancreatic abscess (2), hemi-peritoneum (1). Surgical abstention (3 patients) was determined by: lack of cyst parietal maturation, progressively reduced dimensions and an acute pancreatitis episode. One patient deceased due to haemorrhagic recurrence in tryptic erosion of the portal vein. Surgical decision is supported by US. Surgical tactics shall be adapted to the topography and lesion. Inner derivation represents election indication in mature PKP. In some circumstances, external drainage is however necessary.

Hypothyroidism: its incidence and prevalence in adults older than 55 years of age in an acute rehabilitation unit

OBJECTIVE

To investigate the incidence and prevalence of hypothyroidism in the acute rehabilitation unit.

DESIGN

Retrospective chart review.

SETTING

Inpatient rehabilitation unit.

PATIENTS

Thirty-five men and 91 women older than the age of 55 years (average, 74 years) separated into postsurgical (PS) and nonsurgical (NS) groups. Twenty-two men and 76 women were PS, 21 of whom had a history of hypothyroidism. Thirteen men and 15 women were NS, and 4 in this group had a history of hypothyroidism.

MAIN OUTCOME MEASURES

Levels of thyroid-stimulating hormone and free thyroxine.

RESULTS

There were 34 cases of hypothyroidism, a prevalence rate of 27%. The incidence of newly diagnosed cases was 9% (9 of 101). Six of the newly diagnosed cases were PS patients and three were NS patients. Eleven cases of undertreated hypothyroidism were found, in 9 PS patients and 2 NS patients. The rate of undertreated hypothyroidism in the PS population was 43% (9 of 21); in the nonsurgical population, it was 50% (2 of 4). The overall rate of undertreated hypothyroidism for both PS and NS groups was 44% (11 of 25).

CONCLUSION

There is a high prevalence of hypothyroidism on an inpatient rehabilitation unit and a high rate of undertreated hypothyroidism in PS patients. Screening high-risk patients is recommended.

Comparison between men and women admitted to an inpatient rehabilitation unit after cardiac surgery

OBJECTIVE

To determine whether there are differences in several factors between men and women who undergo inpatient post-cardiac surgery rehabilitation.

DESIGN

A retrospective chart review. Information was collected on a variety of factors: age; previous myocardial infarction; number of days from surgery to admission to rehabilitation; postsurgery, prerehabilitation complications; length of stay on the rehabilitation unit; living arrangements before surgery; disposition; and postdischarge recommendations.

SETTING

Community hospital rehabilitation unit associated with a university hospital.

PATIENTS

One hundred thirty-eight patients (54 men, 84 women) admitted to an inpatient rehabilitation unit after cardiac surgery.

RESULTS

There was a significant relationship between sex and preadmission living arrangements; 56% of women lived alone versus 26% of men (p < .01). There was a statistically significant difference in length of stay on the rehabilitation unit (p < .02). Men stayed longer, with a median stay of 16 days (95% confidence interval, 15 to 20) versus 15 days for women (95% confidence interval, 14 to 15). Ninety-three percent of men were discharged from rehabilitation at 30 days versus 98% of women. No relationship was noted between men and women in age, previous myocardial infarction, number of days from surgery to rehabilitation admission, length of stay on the rehabilitation unit, postsurgery-prerehabilitation complications, complications on the rehabilitation unit, presurgery living arrangements, disposition, and postdischarge therapy recommendations.

CONCLUSION

Men and women showed comparable courses after cardiac surgery. Before surgery, women lived alone more frequently than men.